MECHANISMS OF EPIDERMAL GROWTH FACTOR RECEPTOR ACTIVATION

AFTER EPITHELIAL WOUNDING

by

Ethan Robert Block

B.A. University of California, Santa Cruz, 2001

Submitted to the Graduate Faculty of

The School of Medicine in partial fulfillment

of the requirements for the degree of

Doctor of Philosophy

University of Pittsburgh

2010
 UNIVERSITY OF PITTSBURGH

School of Medicine

This dissertation was presented

by

Ethan Robert Block

It was defended on

January 19, 2010

and approved by

Committee Chair:
Dr. William H. Walker, PhD, Associate Professor, Department of Cell Biology and Physiology

Committee Members:
Dr. Alan H. Wells, MD, DMS, Professor, Department of Pathology

Dr. Sandra A. Murray, PhD, Professor, Department of Cell Biology and Physiology

Dr. Patricia A. Hebda, PhD, Professor, Departments of Otolaryngology and Pathology

Dr. Rebecca P. Hughey, PhD, Professor, Department of Medicine

Dissertation Advisor:
Dr. Jes K. Klarlund, PhD, Associate Professor, Department of Ophthalmology

ii
Copyright by Ethan Robert Block

2010

iii
 MECHANISMS OF EPIDERMAL GROWTH FACTOR RECEPTOR ACTIVATION

AFTER EPITHELIAL WOUNDING

Ethan Robert Block, PhD

University of Pittsburgh, 2010

Wounding disrupts the primary function of an epithelium, which is to provide a barrier to the

outside environment. The longer the epithelial defect remains unhealed, the greater the risks of

morbidity and mortality from infection, loss of tissue homeostasis, and fibrosis. Normally,

epithelial cells restore barrier function by becoming highly motile and migrating to cover the

defect, but in many situations cells do not move fast enough to prevent tissue malfunction. This

is especially true in the cornea, where even minor wounds can impair vision. Therefore, there is

considerable therapeutic interest in identifying signals that induce epithelial migration.

Activation of the epidermal growth factor receptor (EGFR) is a key signaling event that

promotes cells to move and cover wounds in many epithelia. The broad goal of this dissertation

research was to identify mechanisms of wound-induced EGFR activation so that therapies may

be developed to improve normal and pathological healing.

I hypothesized that mechanisms of EGFR activation may differ with respect to distance

from the wound, so I developed wounding models to analyze signaling specifically in cells near

to or far from wounds in a human corneal epithelial cell line. I have examined the

involvement of extracellular ATP, phospholipase D, Src-family kinases (SFKs), and the focal

adhesion kinase Pyk2, all of which are signals that have been hypothesized to be stimulated by

environmental cues related to wounding and to activate the EGFR.

iv
 I have found that the proximal mechanism of EGFR activation is the proteolytic release

of membrane-bound ligands, which is regulated by activation of SFKs. After wounding,

multiple pathways converge on SFKs to regulate EGFR activation and cell motility. In one

pathway, extracellular ATP transactivates the EGFR through phospholipase D2. In a distinct

pathway that functions specifically near the wound edge, Pyk2 triggers SFK and EGFR

activation. Finally, my data suggest the presence of a third distinct pathway that promotes SFK

and EGFR activation in response to a physically unconstrained edge. By delineating signaling

pathways that stimulate EGFR activation, I have identified potential therapeutic targets for

modulating EGFR signaling and cell motility in wound healing and other pathologies.

v
 TABLE OF CONTENTS

PREFACE................................................................................................................................. XVI

1.0 INTRODUCTION........................................................................................................ 1

1.1 EPITHELIAL WOUND HEALING.................................................................. 1

1.1.1 Wound healing in the cornea ....................................................................... 2

1.2 THE EPIDERMAL GROWTH FACTOR RECEPTOR ................................ 7

1.2.1 EGFR signaling and epithelial motility ...................................................... 8

1.2.2 Wound-induced EGFR activation ............................................................... 9

Recently, a number .................................................................................................... 12

1.2.3 EGFR signaling in the clinic ...................................................................... 12

1.3 MODELS TO STUDY WOUND SIGNALING .............................................. 13

1.3.1 The cornea and corneal cell culture .......................................................... 13

1.3.2 Minimizing cell damage.............................................................................. 14

1.3.3 Signaling at the edge ................................................................................... 16

1.4 SUMMARY, GOAL, AND HYPOTHESES ................................................... 19

2.0 DUAL MECHANISMS ACTIVATE THE EGFR AFTER EPITHELIAL

WOUNDING ............................................................................................................................... 21

2.1 INTRODUCTION ............................................................................................. 21

2.2 RESULTS ........................................................................................................... 22

vi
 2.2.1 EGFR activation in cells near the wound edge proceeds independently of

signaling by ATP. ....................................................................................................... 22

2.2.2 ATP functions as a long-range messenger that activates the EGFR...... 24

2.2.3 Wounds heal in the absence of signaling by extracellular ATP.............. 27

2.3 DISCUSSION..................................................................................................... 29

2.3.1 Summary and Model .................................................................................. 32

3.0 ATP TRANSACTIVATES THE EGFR THROUGH PHOSPHOLIPASE D2

SIGNALING................................................................................................................................ 33

3.1 INTRODUCTION ............................................................................................. 33

3.2 RESULTS ........................................................................................................... 34

3.2.1 ATP is released after wounding and activates PLD. ............................... 34

3.2.2 The PLD2 isoform is activated by wounding and by treatment with

ATP.............................................................. 35

3.2.3 Extracellular ATP transactivates the EGFR via PLD2 signaling. ......... 37

3.2.4 PLD2 activation is not required for wound-proximal EGFR activation40

3.2.5 The mechanism of PLD activation by extracellular ATP ....................... 41

3.2.6 PLD2 mediates EGFR activation through SFKs ..................................... 42

3.2.7 Wound-induced cell division depends on signaling by EGFR but not

extracellular ATP. ...................................................................................................... 44

3.3 DISCUSSION..................................................................................................... 46

3.3.1 Summary and model................................................................................... 48

4.0 ATP IS RELEASED PERSISTENTLY FROM WOUNDS AND FREE EDGES

IN EPITHELIAL SHEETS........................................................................................................ 50

vii
 4.1 INTRODUCTION ............................................................................................. 50

4.2 RESULTS ........................................................................................................... 52

4.2.1 ATP is released from cells near edges in epithelial sheets chronically and

independently of damage........................................................................................... 52

4.2.2 A drug screen implicates gap junction hemichannels in ATP release. .. 54

4.2.3 Connexin mimetic peptides do not inhibit ATP release. ......................... 57

4.3 DISCUSSION..................................................................................................... 58

5.0 CHARACTERIZATION OF THE ATP/PLD2-INDEPENDENT MECHANISM

OF WOUND-INDUCED EGFR ACTIVATION: ROLES OF SRC-FAMILY KINASES

AND PYK2 .................................................................................................................................. 61

5.1 INTRODUCTION ............................................................................................. 61

5.2 RESULTS ........................................................................................................... 62

5.2.1 Wounding stimulates EGFR ligand release and EGFR activation in the

absence of extracellular ATP. ................................................................................... 62

5.2.2 Src-family kinases regulate EGFR activation near the wound edge...... 63

5.2.3 Pyk2, but not FAK, is activated rapidly after wounding, independently

of signaling by extracellular ATP, EGFR, or SFK. ................................................ 65

5.2.4 C-Src is activated and associates with Pyk2 after wounding.................. 67

5.2.5 Pyk2 signaling is necessary for SFK and EGFR activation after

wounding..................................................................................................................... 70

5.2.6 Pyk2 activation is necessary for wound healing....................................... 75

5.3 DISCUSSION..................................................................................................... 78

5.3.1 Summary and model................................................................................... 81

viii
6.0 EPITHELIAL FREE EDGES TRIGGER ACTIVATION OF THE EGFR........ 82

6.1 INTRODUCTION ............................................................................................. 82

6.2 RESULTS ........................................................................................................... 83

6.2.1 Chronic EGFR activation is necessary for epithelial migration............. 83

6.2.2 A cell culture model for detecting signaling by free edges. ..................... 85

6.2.3 EGFR activation requires physically unconstrained edges. ................... 86

6.2.4 Free edges stimulate EGFR activation through SFK-mediated

proteolytic ligand release........................................................................................... 90

6.2.5 Free edges stimulate EGFR/ERK activation independently of signaling

by ATP, Pyk2, or loss of polarity .............................................................................. 91

6.3 DISCUSSION..................................................................................................... 94

6.3.1 Summary and model................................................................................... 96

7.0 CONCLUSIONS ........................................................................................................ 97

7.1 A MODEL FOR MECHANISMS OF WOUND-INDUCED EGFR

ACTIVATION .................................................................................................................... 97

7.2 SHORTCOMINGS............................................................................................ 99

7.3 WHAT ARE THE WOUND SENSORS?...................................................... 102

7.3.1 Wound sensors in the extracellular ATP-dependent pathway ............. 103

7.3.2 Wound sensors in the Pyk2 pathway ...................................................... 103

7.3.2.1 Integrin signaling .............................................................................. 103

7.3.2.2 Calcium signaling.............................................................................. 104

7.3.2.3 Reactive oxygen species.................................................................... 106

7.3.3 The free edge sensor.................................................................................. 107

ix
 7.3.4 Redundancy with a purpose..................................................................... 109

7.4 IMPLICATIONS FOR WOUND HEALING THERAPY .......................... 110

7.5 IMPLICATIONS FOR EGFR-RELATED PATHOLOGIES .................... 112

7.5.1 Cancer ........................................................................................................ 112

7.5.2 Cardiac Health .......................................................................................... 113

7.6 SUMMATION.................................................................................................. 114

8.0 MATERIALS AND METHODS ............................................................................ 116

8.1 MATERIALS ................................................................................................... 116

8.2 CELL CULTURE............................................................................................ 117

8.3 WOUND HEALING ASSAY.......................................................................... 118

8.4 WOUNDING MODEL FOR BIOCHEMICAL ANALYSIS OF

SIGNALING IN CELLS NEAR THE WOUND EDGE............................................... 118

8.5 WOUNDING MODEL FOR BIOCHEMICAL ANALYSIS OF

SIGNALING IN CELLS FAR FROM THE WOUND EDGE..................................... 119

8.6 PRODUCTION OF POLYHEMA PLATES WITH PLASTIC STRIPS... 120

8.7 ANALYSIS OF SIGNALING AT VARIOUS DISTANCES FROM FREE

EDGES....................................................................................................................... 120

8.8 WESTERN BLOTTING ................................................................................. 121

8.9 IMMUNOPRECIPITATION ......................................................................... 121

8.10 IMMUNOCYTOCHEMISTRY ..................................................................... 122

8.11 CONFOCAL FLUORESCENCE MICROSCOPY...................................... 122

8.12 EPIFLUORESCENCE MICROSCOPY ....................................................... 122

x
 8.13 GENERATION OF IMAGES CONTAINING BOTH WOUND-

PROXIMAL AND WOUND-DISTAL CELLS ............................................................. 123

8.14 IMMUNOFLUORESCENCE MICROSCOPY OF CELLS ON PLASTIC

STRIPS OR NEXT TO AGAROSE DROPLETS......................................................... 123

8.15 LIVE CELL MICROSCOPY AND DETERMINATION OF CELL

VELOCITIES ................................................................................................................... 124

8.16 RABBIT AND RAT CORNEAL ORGAN CULTURE ............................... 125

8.17 ASSAY FOR PHOSPHOLIPASE D ACTIVITIES ..................................... 125

8.18 AMPHIREGULIN MEASUREMENT.......................................................... 126

8.19 ATP MEASUREMENT .................................................................................. 127

8.20 LUCIFER YELLOW AND RHODAMINE-DEXTRAN UPTAKE ........... 127

8.21 SIRNA TRANSFECTIONS............................................................................ 128

8.22 ADENOVIRAL INFECTIONS ...................................................................... 128

8.23 STATISTICAL ANALYSIS ........................................................................... 129

APPENDIX A ............................................................................................................................ 130

BIBLIOGRAPHY ..................................................................................................................... 132

xi
 LIST OF FIGURES

Figure 1 Anatomy of the eye .......................................................................................................... 3

Figure 2 The cornea has a simple layered structure........................................................................ 4

Figure 3 EGFR activation stimulates signaling cascades that promote motility ............................ 9

Figure 4 Triple Membrane-Passing Signaling (TMPS) mechanism for EGFR activation ........... 11

Figure 5 A method for detecting signaling selectively in wound-distal cells............................... 17

Figure 6 A method for detecting signaling selectively in wound-proximal cells ......................... 18

Figure 7 A method for culturing cells with many free edges ....................................................... 19

Figure 8 A hypothetical model for mechanisms of EGFR activation after epithelial wounding . 20

Figure 9 Wounding induces ATP release ..................................................................................... 23

Figure 10 EGFR activation near the edge is independent of extracellular ATP........................... 24

Figure 11 Extracellular ATP is necessary for EGFR activation near the wound edge................. 25

Figure 12 EGFR and ERK1/2 activation by EGF is not inhibited by apyrase or PLD2 siRNA .. 25

Figure 13 Activation of the EGFR/ERK1/2 pathway is dependent on signaling by ATP in cells

far from, but not near wounds....................................................................................................... 26

Figure 14 Increased velocities of cells near the wound edge are not dependent on extracellular

ATP. .............................................................................................................................................. 27

Figure 15 Extracellular ATP is not necessary for wound healing ................................................ 28

Figure 16 Apyrase was active throughout the healing assays....................................................... 28

xii
Figure 17 RB2 inhibits healing and EGFR activation by AR....................................................... 29

Figure 18 A hypothetical model for the role of extracellular ATP in wound-induced EGFR

activation....................................................................................................................................... 32

Figure 19 ATP released after wounding activates PLD................................................................ 35

Figure 20 Wounding selectively activates the PLD2 isoform ...................................................... 36

Figure 21 Extracellular ATP activates PLD2 ............................................................................... 37

Figure 22 ATP activates the EGFR through PLD2 signaling....................................................... 38

Figure 23 AR is released after wounding or treatment with PA or ATP...................................... 38

Figure 24 AR contributes to EGFR activation after wounding or treatment with PA or ATP..... 39

Figure 25 ATP-stimulated AR release depends on PLD2 ............................................................ 40

Figure 26 EGFR activation near the edge is independent of signaling by PLD ........................... 41

Figure 27 PKC activity is necessary for wound-induced PLD activation and ATP-induced AR

release ........................................................................................................................................... 42

Figure 28 Activation of SFKs is necessary for EGFR transactivation by extracellular ATP....... 43

Figure 29 Wound- and ATP-induced activation of PLD proceeds independently of SFK

activation....................................................................................................................................... 43

Figure 30 PA activates the EGFR through SFK signaling ........................................................... 44

Figure 31 Wound-induced cell division depends on signaling by the EGFR but not extracellular

ATP ............................................................................................................................................... 45

Figure 32 A hypothetical model for the mechanism of EGFR trans-activation by extracellular

ATP after epithelial wounding...................................................................................................... 49

Figure 33 ATP is released persistently after wounding. ............................................................... 53

Figure 34 Cells cultured with many free edges release ATP independently of cell rupture ........ 54

xiii
Figure 35 A drug screen for inhibitors of ATP release................................................................. 55

Figure 36 FFA inhibits the persistent release of ATP after wounding ......................................... 55

Figure 37 ATP release is sensitive to inhibitors of connexons and pannexons ........................... 57

Figure 38 The mimetic peptide Gap26 does not inhibit ATP release........................................... 58

Figure 39 EGFR activation by proteolytic ligand release in the absence of extracellular ATP ... 63

Figure 40 SFKs regulate wound-induced EGFR activation in the absence of extracellular ATP 64

Figure 41 Pyk2, not FAK, is activated rapidly after wounding .................................................... 66

Figure 42 The SFK isoforms c-Src and Fyn were immunoprecipitated from HCLE cells........... 68

Figure 43 C-Src and Fyn are activated after wounding ................................................................ 68

Figure 44 C-Src interacts with Pyk2 after wounding.................................................................... 69

Figure 45 Knockdown of Pyk2 with siRNA inhibits SFK and EGFR activation after wounding 70

Figure 46 Knockdown of Pyk2 with siRNA inhibits SFK and EGFR activation after wounding

but not after stimulation with EGF ............................................................................................... 71

Figure 47 Expression of a Pyk2 dominant-negative mutant inhibits SFK and EGFR activation

after wounding but not after stimulation with EGF ...................................................................... 73

Figure 48 Disruption of Pyk2 signaling by siRNA or PRNK inhibits wound-induced activation of

c-Src and Fyn ................................................................................................................................ 74

Figure 49 Over-expression of wild-type Pyk2 stimulates SFK and EGFR activation ................. 75

Figure 50 Pyk2 activation is necessary for wound healing........................................................... 76

Figure 51 Pyk2 activation is necessary for wound healing........................................................... 77

Figure 52 A hypothetical model for the role of Pyk2 in wound-induced EGFR activation and cell

migration ....................................................................................................................................... 81

Figure 53 Wounding causes persistent EGFR and ERK1/2 activation......................................... 84

xiv
Figure 54 EGFR signaling regulates ERK1/2 activation after wounding..................................... 84

Figure 55 EGFR activity is necessary throughout healing ........................................................... 85

Figure 56 Cells grown on plastic strips have a physically unconstrained edge and slow

proliferation when they reach confluence..................................................................................... 86

Figure 57 Activation of the EGFR by free edges in HCLE cell sheets ........................................ 87

Figure 58 Activation of the EGFR by free edges in MDCK cell sheets....................................... 87

Figure 59 Activation of ERK1/2 by free edges ............................................................................ 87

Figure 60 ERK1/2 activation is limited to near the free edge....................................................... 88

Figure 61 Edges are physically constrained do not activate the EGFR........................................ 89

Figure 62 Free edges activate ERK1/2 through proteolytic release of EGFR ligands ................. 90

Figure 63 Activation of SFKs by free edges is necessary for ERK1/2 activation........................ 91

Figure 64 Free edges activate ERK1/2 independently of extracellular ATP signaling ................ 92

Figure 65 Free edges activate the EGFR independently of signaling by Pyk2 ............................ 92

Figure 66 EGFR ligands are not segregated from receptors......................................................... 93

Figure 67 A hypothetical model for the role of free edges in wound-induced EGFR activation . 96

Figure 68 A working model for mechanisms of EGFR activation after epithelial wounding...... 98

xv
 PREFACE

ACKNOWLEDGMENTS

I am grateful to a number of people for their academic and personal support during my five years

of graduate studies. First of all I thank Dr. Jes Klarlund, who has mentored and advised me not

only during my graduate studies but also in the two years prior, when I worked for him as a lab

tech. I thank him for his constant encouragement, his open-door policy, his firm insistence on

thoughtful planning and high-quality execution, and his willingness to laugh. I recall with some

embarrassment my naivet (not only at the lab bench) when I first came to work for him, and at

this moment of reflection I realize how much of my growth as a scientist I owe to trying to meet

his high standards and garner his trust. I am confident that he has forged in me the proper tools

so that I may, after the next five years, look back on this time and see the same degree of growth.

I owe thanks to a number of other mentors who have guided me over the years. My

dissertation committee, consisting of Patricia Hebda, Rebecca Hughey, Sandra Murray, Alan

Wells, and Will Walker, has dedicated their time and energy into making sure I get the most out

of not only my dissertation research, but also my career. They have contributed solid ideas for

experiments that have aided in manuscript publication and they have challenged me to interpret

data from an un-biased angle, all the while remaining positive and approachable. I owe an extra

thank you to Will Walker for acting as chairperson of the committee and of the Cell Biology

xvi
graduate program. His advice and firm, motivational criticism have aided greatly my research

progress and personal development. I have also benefited from the mentorship of a number of

other scientists whose positive influence I still feel: at the University of Pittsburgh, Q. Jane

Wang, Jon Piganelli, and Jim Funderburgh, and while at UC Santa Cruz, Gummi Thordarson and

Frank Talamantes.

I also thank all of the Klarlund Lab members who have not only contributed to this work

but also made working in the lab all the more enjoyable: Hewa DeFranco, PhD, Chris Guerriero,

PhD, Abigail Mazie, Jennifer Lozano, Julie Spix, Erik Iszkula, Rebecca Sullenberger, and Mike

Tolino. The research presented in Chapter Six was truly a collaborative effort. Specifically,

Mike Tolino contributed figures 54 and 55, and Jennifer Lozano contributed figures 56B

and 64.

The Ophthalmology and Visual Sciences Research Center was the ideal environment for

my graduate studies. I have been exposed to concepts and methodologies in immunology,

virology, biochemistry, biophysics, and medicine that I would not have been had I received

training in the setting of a traditional academic department. I have also had access to the clinical

faculty, and Dr. Francis Mah has graciously allowed me to observe surgeries and clinical visits

so that I could see first-hand the clinical applications of my research. I am grateful to all of the

laboratories and people in the department for courteously helping with reagents and protocols

whenever I came knocking, for encouraging me, and for making the department a fun place to

work. I am especially thankful to Judy Smith, Alice Liang, Lori Snyder, and Katie Nicolette for

doing so much work on all of our behalves behind the scenes. The core imaging module,

directed by Kira Lathrop, is excellent, and this dissertation has greatly benefited from the

microscopy techniques she has helped me to learn.

xvii
 This research has been aided by a number of others who have shared reagents and

facilities: Joseph C. Loftus (Mayo Clinic, Scottsdale, AZ) supplied Pyk2- and PRNK-expressing

adenovirus; Sylvain G. Bourgoin (Universit Laval, Quebec, QC, Canada) supplied anti-PLD2

antibodies; adenovirus encoding tet-OFF was from Ora A. Weisz (University of Pittsburgh,

Pittsburgh, PA); adenovirus encoding GFP was from Michael P. Czech (University of

Massachusetts, Worcester, MA); Ilene K. Gipson (Harvard Medical School, Boston, MA)

supplied the HCLE cells; Bridget M. Deasy and Steven M. Chirieleison (University of

Pittsburgh, Pittsburgh, PA) helped with the time-lapse microscopy. This work was supported by

the National Institutes of Health Grants EY013463, EY08098, and T32 EY017271 and grants

from Research to Prevent Blindness and The Eye and Ear Foundation (Pittsburgh, PA).

It is an honor to thank my family for all they have done for me over the years. My

parents, Bruce and Marian, have given me every opportunity to succeed, but even when I

havent, they have loved and been proud of me. My father-, mother-, and brother-in-law, Ali,

Audrey, and Will, have taught me so much in just a short time, have encouraged me, and have

made me feel that my research is important and exciting. My sister, Adrienne, in whose

footsteps I followed in studying biology, has given me unyielding support, endless reasons to

laugh, and the inspiration to do good in my life. My wife, Azi, aka Dr. Block, PhD, has simply

saved my life. Her energy, intelligence, work ethic, compassion, and love have given me the

strength to persevere, the motivation to succeed, and the confidence to believe in myself.

Finally, I thank our two-month-old son, Jonah, for filling my life with joy. I have never before

looked so positively toward the future. To all of you, my family, I dedicate this dissertation.

xviii
 ABBREVIATIONS

ADAM A disintegrin and metalloprotease

AG Tyrphostin AG 1478 (EGFR inhibitor)

APY Apyrase

AR Amphiregulin

CBX Carbonexolone

Cx Connexin

DAG Diacylglycerol

EGF Epidermal growth factor

EGFR Epidermal growth factor receptor

ErbB Erythroblastic leukemia viral oncogene homolog

ERK Extracellular signal-regulated kinase

FAK Focal Adhesion Kinase

FFA Flufenamic acid

GA Glycerrhetinic acid

GPCR G-protein-coupled receptor

HB-EGF Heparin binding EGF-like growth factor

HCLE Human Corneal Limbal Epithelial

HGF Hepatocyte growth factor

LASIK Laser-assisted in situ keratomileusis

xix
LDH Lactate dehydrogenase

LPA Lysophosphatidic acid

LY Lucifer yellow

MAPK Mitogen-activated protein kinase

MARCKS Myristoylated alanine-rich C kinase substrate

MMP Matrix metalloprotease

P2Y Type 2-Y purinergic receptor

PA Phosphatidic acid

PBS Phosphate-buffered saline

PIP2 Phosphatidylinositol (4,5)-bisphosphate

PKC Protein Kinase C

PLC Phospholipase C

PLD Phospholipase D

polyHEMA Poly(2-hydroxyethyl methacrylate)

PP2 Src-family kinase inhibitor PP2

PRK Photorefractive keratectomy

PRNK Pyk2-related non-kinase

Pyk2 Proline-rich tyrosine kinase 2, or focal adhesion kinase 2

RB2 Reactive Blue 2

SFK Src-family kinase

SKI Src-family kinase inhibitor I

TGF Transforming growth factor

xx
 1.0 INTRODUCTION

1.1 EPITHELIAL WOUND HEALING

The fundamental role of an epithelium is to provide a barrier to the outside environment.

Wounding severely disrupts the epithelial barrier, leaving the organism prone to infection, fluid

loss, and other consequences of unregulated permeability. The longer a wound takes to heal, the

greater the risk to life and limb. Although powerful and efficient mechanisms have evolved to

restore barrier function rapidly after wounding, many wounds heal inefficiently, resulting in

malfunctioning tissue, and some wounds do not heal at all.

Wounds come in a variety of forms: wounds to the external epithelia such as the skin

and the cornea can be caused by traumatic abrasions, chemical burns, or surgery; wounds to the

internal epithelia such as the gastrointestinal tract and the airway are typically caused by

inflammatory diseases. Genetic, molecular, and cellular approaches have shown that wound

repair is accomplished through a complex series of events that may involve varying degrees of

participation by not only epithelial cells, but also fibroblasts in the underlying stroma, platelets,

nerves, immune and inflammatory cells, and endothelial cells (Cole et al., 2001; Grose and

Werner, 2004; Gurtner et al., 2008; Jacinto et al., 2001; Jane et al., 2005; Martin, 1997).

The role of non-epithelial cells in the healing process varies greatly depending on the

cause and location of the wound, but the normal response of epithelial cells themselves is quite

1
similar: cells become highly motile and migrate as a collective sheet to close the epithelial defect

(Friedl and Gilmour, 2009; Martin and Parkhurst, 2004). Epithelial cells do eventually divide to

replace lost cells, but the quickest fix to re-epithelialization is cell migration.

When an epithelium is wounded, cells at the edge undergo profound phenotypic changes

and subsequently acquire an ability to migrate rapidly. Cell motility is a complex multistep

process that is accomplished through repeated cycles of 1. protrusion of the leading edge into the

free space, 2. adhesion of these protrusions to the extracellular matrix, 3. de-adhesion from the

matrix at the trailing edge, and 4. contraction of the cell body. This biomechanical feat is

accomplished through highly integrated processes involving the modulation of substrate

adhesions and rearrangements of the dynamic actin cytoskeleton (Jacinto et al., 2001; Kaverina

et al., 2002; Lauffenburger and Horwitz, 1996; Mitchison and Cramer, 1996; Nobes and Hall,

1999; Thiery and Sleeman, 2006).

The mechanics of cell motility during wound healing have been studied extensively, but

the cellular signals that induce these dramatic phenotypic changes after wounding are poorly

understood. The broad goal of this dissertation research was to clarify the cellular signals that

promote the induction of motility after wounding so that therapies may be developed to

ameliorate normal and pathological healing.

1.1.1 Wound healing in the cornea

Of all epithelial wounds, those inflicting the cornea are perhaps the most prone to causing major

morbidity, even for small wounds in otherwise healthy individuals. The reason for this is simply

anatomical (Figure 1). The cornea is the most anterior tissue in the eye; it is the first tissue

through which light travels on its path through the lens and finally to the retina, where

2
photostimuli are transduced into neural impulses. The cornea is normally transparent and is

responsible for roughly two thirds of the eyes refractive power. The integrity and health of the

cornea are critical for the maintenance of transparency, so wounding can lead to low vision and

even blindness. Corneal pathologies are the second leading cause of blindness worldwide and

complications from corneal wounds account for an estimated one to two million new cases of

monocular blindness every year (Whitcher et al., 2001).

Figure 1 Anatomy of the eye

The cornea is the most anterior tissue and is the first to refract light

as it travels to the retina. The limbus (not marked) is the area

where the cornea meets the sclera.

The transparency and refractive power of the cornea are a result of the highly organized

collagen/proteoglycan lattice that comprises the corneal stroma. As seen in the human corneal

cross-section in figure 2, the stromal layer makes up about 90% of the corneas thickness. It is

interspersed with a syncytium of specialized mesenchymal cells, keratocytes, which regulate the

production and organization of the collagen lattice (Funderburgh, 2002; Funderburgh et al.,

2003). The most anterior layer of the cornea is a stratified layer of corneal epithelial cells. The

epithelium is bathed in the tear film, which contains mucous, aqueous (ions, growth factors, etc),

and lipid components (Gipson, 2007). An endothelial monolayer separates the stroma from the

aqueous humor and regulates hydration and nutrition of stromal components. Aiding

3
transparency even further, the cornea is normally avascular and plays host to very few resident

immune cells. The cornea is, however, heavily innervated. These nerves secrete neurotropic

factors such as substance P that support the health of epithelial cells and their stimulation signals

tearing responses by the lacrimal glands (Davis and Dohlman, 2001; Tsubota, 1998).

Figure 2 The cornea has a simple layered

structure

The three cellular layers of the cornea: the

stratified epithelium, the stroma with

interspersed keratocytes (dark spots), and the

endothelial monolayer. The cornea is

normally avascular and heavily innervated (not

visible).

Epithelial wounds and defects heal through tightly regulated and highly coordinated

phenotypic changes to the corneal cell layers (Bazan, 2005; Dupps and Wilson, 2006; Fini and

Stramer, 2005; Netto et al., 2005; Wilson et al., 2001; Wilson et al., 2003). Immediately after

wounding, cytokines released from wounded epithelial cells such as interleukin-1 and tumor

necrosis factor- induce apoptosis of stromal keratocytes proximal to the sight of injury. This

may serve as a safety mechanism for inhibiting the replication of infectious agents.

Simultaneously, in the epithelium, growth factors, such as epidermal growth factor (EGF) and

hepatocyte growth factor (HGF) in the tear film begin to stimulate cell motility and re-

epithelialization (Baldwin and Marshall, 2002; Imanishi et al., 2000; Klenkler and Sheardown,

2004; Klenkler et al., 2007; Yu et al., 2009).

4
 As re-epithelialization progresses, the corneal stroma commences repair processes of its

own. Immune cells are summoned by chemokines to clear debris and fight infection. Also, the

keratocytes in the stroma migrate to the wound site and undergo a dramatic phenotypic

transformation into myofibroblasts, in large part due to the actions of transforming growth

factor- (Saika, 2004). Myofibroblasts remodel and repair the collagen lattice and promote

migration of epithelial cells.

While these repair processes are necessary for proper healing, they are also potentially

disastrous to vision. Excess numbers of immune cells and myofibroblasts can impair vision

because they are less transparent than keratocytes. Also, excessive stimulation of immune cells

can lead to further inflammation and even corneal vascularization. Excessive myofibroblast

stimulation can lead to the deposition of disorganized extracellular matrix (fibrosis), leading to

corneal haze or scaring (Netto et al., 2006). The signals that can hyper-stimulate the immune

and keratocyte responses are present as long as re-epithelialization is incomplete (Netto et al.,

2005), so clearly, the faster epithelial cells migrate to cover the wound, the lesser the chance of

poor clinical outcomes.

Causes and clinical outcomes of corneal wounds are highly variable. Accidental

abrasions that do not penetrate the epithelial basement membrane typically heal rapidly and

without complication (Wilson and Last, 2004). Wounds that penetrate into the stroma or through

the endothelium take longer to heal and are therefore more prone to infection and scarring. The

cornea can also be wounded by chemical, thermal, and electrical burns, which are further

complicated by persistent modification of the basement membrane that can lead to corneal ulcers

(Kato et al., 2006).

5
 Elective vision-correcting surgeries represent a class of increasingly common corneal

wounds (Kuo, 2004; Schallhorn et al., 2006). During laser-assisted in situ keratomileusis

(LASIK) an incision is made around almost the entire circumference of the cornea and this

corneal flap is lifted to allow correction of the stromal surface beneath. Epithelial cells must heal

around the edges when the flap is replaced. During photorefractive keratectomy (PRK), the

underlying stroma is accessed by complete debridement of the central cornea, and peripheral

epithelial cells must migrate to cover the denuded stroma.

A number of underlying medical conditions can also cause corneal defects, which are

similar to wounds in that they result in the breakdown of the epithelial barrier due to loss of

epithelial cells. Some of the most common of these include viral infection (Kaye and

Choudhary, 2006), fungal infection (Thomas, 2003), dry eye (the most common presenting

ocular disorder that is a collection of diseases affecting the volume and/or composition of the

tear film) (Gipson, 2007; Tsubota, 1998) and neurotropic keratitis (the loss or disease of corneal

nerves) (Bonini et al., 2003; Nishida and Yanai, 2009; Pushker et al., 2001).

Not only do these underlying conditions cause epithelial defects, but along with diabetes

(Chikama et al., 2007), prior corneal transplant (Obata and Tsuru, 2007), and corneal dystrophies

(Das et al., 2005) they also impair healing of even minor surface abrasions. The damage to

corneal nerves from LASIK and PRK causes defects in tears and neurotropic factors that

impedes re-epithelialization not only during recovery from surgery but also after future wounds

(Belmonte, 2007).

The longer a corneal wound or defect remains open, the higher the risks of scaring,

infection, and loss of sight, yet there is currently no commonly-used therapy for increasing

healing rates of the corneal epithelial cells (Cameron, 2007; Duane et al., 2002; Wilson and Last,

6
2004). Treatment of corneal wounds consists of antibiotics, and when applicable, treatment of

the underlying corneal pathology. The great clinical need for therapies that increase epithelial

healing rates has inspired investigators for decades to study signal transduction pathways that

regulate cell motility and to apply their findings in the clinic.

1.2 THE EPIDERMAL GROWTH FACTOR RECEPTOR

One molecular signal that has been identified as being important for epithelial motility is

activation of the EGF receptor (EGFR, also ErbB1) (Jorissen et al., 2003; Warren and Landgraf,

2006; Wells, 1999). Activation of the EGFR tyrosine kinase is stimulated by wounding and is

absolutely required for enhanced cell migration and healing of numerous epithelia, including the

gut, airway, epidermis, and cornea (Goodlad and Wright, 1995; Puddicombe et al., 2000;

Repertinger et al., 2004; Zieske et al., 2000).

The EGFR is a receptor tyrosine kinase of the ErbB family that also includes ErbB2

(HER2/Neu), ErbB3, and ErbB4. Unliganded ErbB receptors are held in an autoinhibited

conformation, but upon ligand binding, allosteric conformational changes release autoinhibition,

increase receptor dimerization, and elevate kinase activity (Zhang et al., 2006). Trans- and

autophosphorylation of cytoplasmic tyrosine residues after activation can be detected by Western

Blotting. I routinely use antibodies directed against the EGFR phosphorylated on tyrosine-1173

as a marker for receptor activation.

7
1.2.1 EGFR signaling and epithelial motility

The tyrosine-phosphorylated EGFR activates a myriad of signaling cascades that promote

epithelial cell motility (Figure 3). Phosphotyrosine (pY) residues serve as docking platforms for

numerous substrates, enzymes, and adaptors. For example, pY1173 is the docking site for

phospholipase C (PLC)-, which hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP2) and

stimulates calcium signaling, as well as for the adaptor protein Shc, which couples EGFR

activation to signaling by the mitogen-activated protein kinases (MAPKs) Extracellular signal-

Regulated Kinases (ERK1/2, or p44/42 MAPK) (Katz et al., 2007; McKay and Morrison, 2007;

Rubinfeld and Seger, 2005). Other pY residues stimulate activation of phosphatidylinositol 3-

kinase and Src kinase signaling (Jorissen et al., 2003; Yarden and Sliwkowski, 2001).

The myriad of signaling cascades activated downstream of the EGFR work in concert to

stimulate motility through a number of effector pathways that directly regulate the biomechanics

of motility. Importantly, EGFR activation, at times through ERK1/2, leads to the dissolution of

cell/matrix bonds that stabilize epithelial cells in place and mobilizes actin-based protrusions and

contractions that propel the cell forward (Chen et al., 1996; Dieckgraefe et al., 1997; Dise et al.,

2008; Fitsialos et al., 2007; Glading et al., 2000; Hudson and McCawley, 1998; Jorissen et al.,

2003; Wells et al., 1998; Xie et al., 1998).

8
Figure 3 EGFR activation stimulates signaling cascades that promote motility

An EGFR ligand binds the receptor leading to receptor dimerization and transphosphorylation of cytoplasmic

domain tyrosines. Phosphotyrosines present docking sites for enzymes and adaptors that initiate numerous signaling

cascades, including ERK1/2, that promote motility through various effector pathways. Depicted as a unidirectional

linear pathway for simplicity. See text for details.

1.2.2 Wound-induced EGFR activation

ErbB ligands such as amphiregulin (AR), heparin-binding EGF-like growth factor (HB-EGF),

and transforming growth factor- (TGF) are synthesized as membrane-bound precursors

9
(Harris et al., 2003; Higashiyama et al., 2008). The EGFR can be activated by a variety factors

that stimulate the release of these ligands, which is dependent on proteases of the matrix

metalloprotease (MMP) and A Disintegrin And Metalloprotease (ADAM) families (Blobel,

2005; Daub et al., 1996; Fischer et al., 2003; Ohtsu et al., 2006; Sanderson et al., 2006). EGFR

activation by proteolytic ligand release, first described following stimulation with G-protein-

coupled receptor (GPCR) agonists, has been termed the triple membrane-passing signaling

(TMPS) mode of EGFR activation and is now known to occur following stimulation by a variety

of stimuli such as ultraviolet radiation and receptor tyrosine kinase activation (Figure 4)

(Higashiyama et al., 2008; Prenzel et al., 2000).

The proximal event for wound-induced EGFR activation in numerous epithelia such as

the skin, gut, airway, and cornea is this very TMPS mechanism (Block et al., 2004; Davies et al.,

1999; Myhre et al., 2004; Shirakata et al., 2005; Tokumaru et al., 2000; Xu et al., 2004).

Methods to study the TMPS mechanism include the use of inhibitors of the ligand proteases,

anti-ligand neutralizing antibodies, anti-EGFR antibodies that bind to and block the ligand

binding domain, and conditional knockouts of EGFR ligands. In the corneal epithelium, EGFR

activation and the induction of cell motility after wounding is regulated by the ligands HB-EGF

and AR (Block et al., 2004; Mazie et al., 2006; Spix et al., 2007; Xu et al., 2004).

The intracellular signals that regulate activation of EGFR ligand proteases are poorly

understood, but recent data suggest the involvement of Src-family kinases (SFKs) (Zhang et al.,

2004). The Src family includes the ubiquitously expressed c-Src, Fyn, and Yes isoforms. These

kinases regulate numerous cellular events related to cytoskeletal dynamics, cell/matrix

interactions, and the transduction of extracellular signals (Abram and Courtneidge, 2000; Ingley,

2008; Roskoski, 2005; Thomas and Brugge, 1997). It is likely that SFKs also control EGFR

10
ligand shedding after wounding because they are activated after wounding and their activity is

necessary for EGFR activation (Xu et al., 2006; Yamada et al., 2000). Therefore, deciphering

mechanisms of wound-induced EGFR activation may go hand-in-hand with understanding

mechanisms of SFK activation.

Figure 4 Triple Membrane-Passing Signaling (TMPS) mechanism for EGFR activation

Extracellular stimuli such as a GPCR agonist or UV irradiation trigger activation of poorly understood intermediate

signals that include SFKs. Consequently, a protease of the MMP or ADAM family is activated, which cleaves an

EGFR pro-ligand. The released ligand then binds and activates the EGFR.

Signaling by extracellular ATP, by phospholipase D (PLD), and by the focal adhesion

kinase Pyk2, among others, have been recently hypothesized to contribute to wound-induced

EGFR activation because they are known to be active after wounding and to stimulate SFK

and/or EGFR signaling in other systems (Boucher et al., 2007; Klepeis et al., 2004; Mazie et al.,

2006; Yang et al., 2004; Yin et al., 2007; Yu et al., 2009)

11
 Recently, a number

1.2.3 EGFR signaling in the clinic

The overwhelming experimental evidence indicating that EGFR activation regulates wound

closure suggests an obvious and simple clinical solution for corneal wounds: treatment with

EGF. The actual clinical implementation and success of such treatments, however, has proven to

be not so simple. Whereas a number of studies have indicated that EGF accelerates corneal

wound healing in a variety of animal models and in humans (Baldwin and Marshall, 2002;

Imanishi et al., 2000; Klenkler and Sheardown, 2004; Lu et al., 2001; Pastor and Calonge, 1992),

the overall clinical utility of EGF has been hampered by a number of problems. The biggest

hurdle to EGF therapy has been one of drug delivery. EGF and most soluble, non-cell-

permeable factors are rapidly washed away from the epithelial surface in the tear film after

blinking, so frequent high doses, exceeding five applications of 10 g/ml EGF solutions per day,

have been required. Another complication with EGF therapy has been activation of

myofibroblasts in the corneal stroma, which is exposed after wounding. Activated

myofibroblasts can cause low vision by promoting corneal haze through the aberrant production

and distribution of extracellular matrix (He and Bazan, 2008). As a result of these

complications, EGF is not commonly used and, as stated earlier, there remains no established

therapy for hastening the corneal healing process (Cameron, 2007).

The broad goal of this dissertation was to identify molecular mechanisms of wound-

induced EGFR activation. It is expected that such information will allow new therapeutic

strategies to be developed for improving normal and pathological healing of many tissues. For

example, the identification of factors upstream of proteolytic ligand release could suggest

therapies that overcome many of the hurdles encountered by EGF therapy by promoting

12
increased and long-lasting shedding of endogenous EGFR ligands. Additionally, delineating

signaling pathways leading to EGFR activation may help to identify mechanisms by which

conditions such as diabetes, infection, and dry eye inhibit healing (Kaye and Choudhary, 2006;

Klenkler et al., 2007; Wakuta et al., 2007).

1.3 MODELS TO STUDY WOUND SIGNALING

1.3.1 The cornea and corneal cell culture

For decades, the cornea has been a popular model to study wound healing. The first reason for

this is clinical: as detailed in the previous section, it is necessary to understand the cellular

mechanisms of corneal epithelial wound healing to treat wounds and improve surgical outcomes.

The second reason is practical: the cornea is easily accessible for study in intact animals as well

as in organ culture (of whole globes or excised corneas) and in cell culture. The final reason is

that the corneal epithelium serves as a model for epithelial cell motility in general. The cornea

has a simple layered structure (Figure 4) and is normally avascular and wounds heal by the

collective migration of epithelial sheets. One can apply what is learned from corneal healing,

which is typically efficient and scar-free, to the normal and pathological motility of many

epithelia.

Cell culture is a relatively cost-efficient yet powerful method for studying epithelial

healing. Although some aspects of wound healing are lost, such as interactions with vascular

and lymphatic elements, the most basic and essential aspect of wound healing, that of epithelial

migration, can be closely approximated in cell culture. The signals that stimulate a wounded

13
epithelial sheet to migrate have been studied extensively in cell culture. Many growth factors,

including EGF, were observed to enhance migration of cultured cells, leading to the development

of promising therapies (Hori et al., 2007; Imanishi et al., 2000; Pastor and Calonge, 1992;

Watanabe et al., 1987).

Cultured primary corneal epithelial cells and cell lines have been used extensively.

Human corneal epithelial cells grown from primary tissue explants (Ebato et al., 1987) most

closely resemble cells in vivo, but donor tissue is scarce and highly variable due to age and

genetic background. Primary cells grown from rabbit explants are less scarce and less variable,

but are incompatible with many human-specific reagents. Human corneal epithelial cells

immortalized by the SV40 large T-antigen have been established and are frequently used (Ali

and DeCaprio, 2001; Kahn et al., 1993). Most of the experiments described here have been

performed in a different cell line, Human Corneal Limbal Epithelial (HCLE) cells, which have

been immortalized by abrogation of p16INK4A/Rb and p53 functions and overexpression of the

catalytic subunit of the telomerase holoenzyme (Gipson et al., 2003). They express markers of

the differentiated corneal epithelium such as the corneal epithelial-specific keratins K3 and K12,

and serve as a model for native corneal epithelial cells.

1.3.2 Minimizing cell damage

Wound healing has traditionally been studied in cultured cells by use of the scratch assay (Liang

et al., 2007): a confluent cell sheet is scratched with a pipet tip or similar implement, scraping

away a swath of cells, and cell migration is monitored over time. In response to the scratch,

epithelial cells migrate into the denuded area as a collective sheet. The scratch assay has proved

14
to be an excellent method for studying collective cell migration, a process distinct from

unicellular (amoeboid or mesenchymal) migration (Friedl, 2004; Friedl and Gilmour, 2009).

To study the cellular signaling caused by wounding, the scratch assay has again been an

effective tool. Following singular or repeated scratches, signals are detected either by various

imaging techniques or by preparing extracts of the remaining cells for biochemical analysis.

Scratch-wounding of cultured cells closely approximates an in vivo scratch wound and is an

essential tool for understanding the breadth of epithelial responses to wounding.

A scratch wound is a complex stimulus: cells are challenged with the loss of cell/cell

contacts, cell stretch, cell lysis, and the sudden availability of free space, among other

perturbations. Each of these factors stimulates a host of cellular signals that regulate the

numerous epithelial responses to wounding: cells rupture and die, they undergo apoptosis, they

repair damaged membranes, they proliferate, they increase phagocytosis of cell debris, and they

increase production of extracellular signals that affect inflammation or vascularization. To

complicate matters, these processes may vary greatly depending on the tissue or the type of

wound.

Of all the epithelial responses to wounding, the only one that is absolutely required for re-

epithelialization is the induction of motility. Identifying the precise cues that induce motility

after wounding is a major focus of wound research in general (Block et al., 2004; Martin and

Parkhurst, 2004; Tschumperlin, 2004; Vermeer et al., 2003; Xu et al., 2006; Yin et al., 2007),

and of this dissertation in particular. From a practical standpoint, it is therefore necessary to

eliminate some of the signaling noise caused by wounding so that only the signals that regulate

the transition to motility can be studied.

15
 To minimize the signaling effects of cell damage and the loss of cell/cell contacts, we

have previously developed a method for wounding sheets of cells by the removal of agarose

barriers: cells are cultured to confluence around a thin agarose strip to which they do not adhere,

and the cell sheet is subsequently wounded by the removal of this strip. We do not detect cell

damage after wounding cell sheets in this manner, and importantly, gaps close at the same rate as

when wounds are made by scraping (Block et al., 2004). Similar observations have since been

made using a thin film of polydimethylsiloxane instead of agarose (Poujade et al., 2007).

Therefore, the signals that are necessary for normal wound healing in culture are elicited simply

by the removal of physical barriers.

I use the term wounding to describe the stimulation of cell sheets by the removal of

agarose barriers, but I recognize that it is somewhat misleading, given that there is no (or

minimal) actual damage to the cells. Since the removal of physical barriers stimulates cells with

a subset of the stimuli induced by conventional wounding while promoting motility to a similar

degree, I feel that this research is directly applicable to wound healing and that the terminology

is at least less cumbersome, if not justified.

1.3.3 Signaling at the edge

A number of studies have reported variations in cell signaling with respect to distance from the

wound. For example, wound-induced increases in intracellular calcium concentrations near the

wound edge depend on extracellular calcium, but increases far from the wound edge depend on

intracellular calcium (Klepeis et al., 2001). Also reactive oxygen species have been observed in

cells near the wound edge but not farther away (Nikolic et al., 2006). This may be explained by

the obvious hypothesis that cells near the edge of a wound are subjected to a number of spatial

16
and mechanical perturbations while cells far from a wound are more influenced by the release of

soluble factors. Many studies also assert that the cells near the wound edge are vital for healing

because it is in edge cells where activation of the cellular migration machinery is necessary

(Fenteany et al., 2000) and where increased velocities have been measured (Farooqui and

Fenteany, 2005; Poujade et al., 2007). It is therefore important to study spatial aspects of

signaling and to characterize signaling specifically in near-edge cells.

To monitor the spatial aspects of EGFR signaling in my own studies, I have developed

methods to biochemically analyze cells that are either near or far from wounds. To prepare

extracts of cells far (>3 mm) from wounds for biochemical analysis, reactions are slowed on ice

before removing cells near the wound, and then prepare extracts from the cells that remain

(Figure 5 and for details see Methods 8.5). As a control, cells are scraped away from a

confluent, unwounded culture.

Figure 5 A method for detecting signaling

selectively in wound-distal cells

Agarose is shown in pink. After stimulation, 3

mm from each side of the wound is removed.

Due to low yield, the cells that are removed from near the wound are difficult to analyze.

To selectively analyze cells near the wound edge, we have previously developed a method for

culturing cells around agarose droplets, and then wounding the cells by removing the droplets

17
(Block et al., 2004) (Figure 6 and for details see Methods 8.4). As a control, agarose droplets are

left in place or removed on ice.

Figure 6 A method for detecting

signaling selectively in wound-proximal

cells

Agarose is shown in pink. Phase contrast

image is of cells grown around agarose

droplets.

Wounding by the removal of agarose droplets minimizes cell damage in comparison to

scrape-wounding (see Chapter Four), but cells are still stimulated by numerous factors such as

free extracellular matrix and the loss of physical constraints. Because physically unconstrained,

or free edges are inherently present in all wounds, we have also developed a model to study the

signaling caused by the presence of a free edge in the absence of cell damage or leading-edge

matrix adhesions. In this model, cells are cultured on thin plastic strips that sit on polyHEMA, a

compound to which cells do not adhere (Figure 7 and for details see Methods 8.6). Cultures

prepared on plastic strips have many free edges, while control cultures have virtually none. A

major difference between this and the agarose methods is that cells cultured on plastic strips are

not stimulated acutely. Therefore, the plastic strip method may be more applicable to studying

the signaling from cells in migrating epithelial sheets, which occur hours after wounding or

during development.

18
 Figure 7 A method for culturing cells with many free edges

Schematic of plates covered with polyHEMA and plastic strips

(not to scale). Light grey is polyHEMA; dark grey is plastic.

Phase contrast image is of cells grown on plastic strips.

The models described above minimize signaling noise, yet still allow the study of signals

that induce migration. Additionally, these models allow me to determine the spatial range of a

signal and to study signaling specifically at the wound edge.

1.4 SUMMARY, GOAL, AND HYPOTHESES

The function of a wounded tissue and the health of an organism is jeopardized as long as an

epithelial wound remains open. Wounds to the cornea can lead to impairment or even loss of

vision. Cell migration is absolutely required for the closure of epithelial wounds, and EGFR

activation has been identified as an essential signal for the induction of motility after wounding.

To study EGFR activation, I have developed novel wounding models that minimize cell damage

and allow biochemical analysis of cells specifically near the wound edge.

The goal of this dissertation was to gain a better understanding of the cellular signals that

induce migration after epithelial wounding. By clarifying mechanisms of EGFR activation after

wounding, my findings are expected to suggest new therapeutic avenues for improving normal

and pathological healing. To that end, I have tested the following primary hypotheses (and see

diagram in Figure 8):

19
1. Mechanisms of EGFR activation vary with respect to distance from a wound.

2. Extracellular ATP stimulates wound-induced EGFR activation through Phospholipase D.

3. ATP is released from the wound edge through connexin hemichannels.

4. EGFR activation near the wound edge is regulated by Src-family kinases and Pyk2.

5. EGFR activation is stimulated by the presence of a free edge.

To distal cells

Ligand
Ch. 2
MMP EGFR
ADAM
P2R

ATP Ch. 3
Ch. 4
PLD2 PA

Ch. 6
Free edge SFK
Ch. 5 Motility
Sensor(s) Pyk2

Figure 8 A hypothetical model for mechanisms of EGFR activation after epithelial wounding

A number of hypotheses have been tested in this dissertation to clarify mechanisms of wound-induced EGFR

activation. The chapters describing the testing of these hypotheses are shown near their respective pathways. P2R,

type-2 purinergic receptor; PLD2, phospholipase D2; PA, phosphatidic acid; Pyk2, the focal adhesion kinase Pyk2;

SFK, Src-family kinase; MMP, matrix metalloprotease; ADAM, a disintegrin and metalloprotease; EGFR,

epidermal growth factor receptor.

20
 2.0 DUAL MECHANISMS ACTIVATE THE EGFR AFTER EPITHELIAL

WOUNDING

2.1 INTRODUCTION

Activation of the EGFR is absolutely required for the induction of epithelial motility after

wounding. Recent research suggests that extracellular ATP may regulate wound-induced EGFR

activation (Boucher et al., 2007; Yin et al., 2007). In addition to their numerous intracellular

functions, ATP and other nucleotides function as extracellular signaling molecules (Burnstock,

2006; Schwiebert and Zsembery, 2003). Extracellular ATP signals by binding to the P2X class

of cation channels and the P2Y class of G-protein coupled receptors (Abbracchio et al., 2006).

ATP may also be metabolized by ecto-enzymes to adenosine, which activates the P1 class of

receptors. ATP is released from epithelial cells following mechanical perturbation and is

released from corneal epithelial cells after wounding (Boucher et al., 2007). Treatment with

extracellular ATP promotes EGFR activation and healing of wounds in corneal epithelial cells

through P2Y receptor activation (Boucher et al., 2007; Klepeis et al., 2004; Yang et al., 2004;

Yin et al., 2007).

In this section, I have tested the hypothesis that signaling by extracellular ATP is

necessary for wound-induced EGFR activation. Furthermore, since extracellular ATP is freely

diffusible, I was interested in analyzing the spatial aspects of EGFR activation in wounded

21
epithelial cell sheets. Numerous studies have concluded that cell signaling and phenotype vary

in relation to distance from the wound (Fenteany et al., 2000; Klepeis et al., 2001; Nikolic et al.,

2006; Omelchenko et al., 2003; Poujade et al., 2007), and that the major driving forces for

epithelial migration appear to be derived mainly from the first few rows of cells from the wound

edge (Farooqui and Fenteany, 2005; Fenteany et al., 2000). Therefore, I determined whether the

signals that induce EGFR activation vary as a function of distance from the wound edge, and

whether ATP signaling is necessary for the induction of cell motility following wounding.

Portions of this chapter were published previously, ASCB copyright 2008, in

http://www.molbiolcell.org/cgi/content/full/19/11/4909 (Block and Klarlund, 2008) and have, in

some cases, been modified.

2.2 RESULTS

2.2.1 EGFR activation in cells near the wound edge proceeds independently of signaling

by ATP.

Mechanically wounding sheets of epithelial cells results in robust release of ATP, presumably

mostly as a result of leakage from damaged cells (Klepeis et al., 2001; Yang et al., 2004; Yin et

al., 2007). To test whether ATP is released by our wounding model of removing agarose

droplets (see Chapter One, Figure 5 and Methods), I collected conditioned media from cells

wounded by this procedure and performed an ATP bioluminescence assay. As is seen in Figure

9, wounding by this procedure also results in significant release of ATP.

22
 5 Figure 9 Wounding induces ATP release
4
mg protein
nmol ATP/

ATP contents in conditioned media were determined in HCLE cell cultures

3
2 that were unwounded (NT), wounded and incubated with medium for 10
1 minutes (W), or treated similarly in the presence of 5 U/ml apyrase (W+APY).
0
NT W W+APY

To test whether extracellular ATP is necessary for wound-induced EGFR activation, I

wounded HCLE cells in the presence of apyrase, which dephosphorylates extracellular ATP and

ADP to AMP. AMP is not known to signal, but can be further metabolized to adenosine, which

signals through P1 purinergic receptors. To assess activation, I performed Western Blotting for

the phosphorylated and total forms of the EGFR and of Extracellular signal-Regulated Kinases 1

and 2 (ERK1/2), which are down-stream targets of the EGFR that are important for wound-

healing (Dieckgraefe et al., 1997; Fitsialos et al., 2007; Glading et al., 2000; Katz et al., 2007;

McKay and Morrison, 2007; Rubinfeld and Seger, 2005). Although apyrase reduced ATP

concentrations in the conditioned medium to below detectable limits (Figure 9), wound-induced

activation of EGFR and ERK1/2 were unaffected by apyrase (Figure 10A). Extracellular ATP is

known to stimulate protein kinase C (PKC) activity by binding P2Y and P2X purinergic

receptors at the cell surface (Burnstock, 2006; Schwiebert and Zsembery, 2003). Therefore, as a

further control to verify that the apyrase was functional, I immunoblotted for the phosphorylated

forms of Myristoylated Alanine-Rich C-Kinase Substrate (MARCKS), an indicator of PKC

activity (Aderem, 1992) and observed that PKC activation was indeed blocked by apyrase

(Figure 10A).

EGFR activation after wounding depends on the proteolytic release of ligands such as

amphiregulin (AR) and HB-EGF (Block et al., 2004; Xu et al., 2004). I was able to detect

23
 Figure 10 EGFR activation near the edge is

independent of extracellular ATP

A. Western Blots for activation of EGFR,

ERK1/2 and MARCKS after no treatment

(NT), wounding (W), and/or treatment with 30

U/ml Apyrase (APY). B. Quantities of AR in

conditioned media from cells treated as in A.

AR, but not HB-EGF, by ELISA of media conditioned by HCLE cells. AR release from cells

near the edge was stimulated by wounding, and this was not affected by treatment with apyrase

(Figure 10B). These results indicate that EGFR and ERK1/2 activation proceeds independently

of extracellular ATP in cells near the wound edge.

2.2.2 ATP functions as a long-range messenger that activates the EGFR.

Because ATP is freely diffusible, I next examined whether extracellular ATP mediates EGFR

activation in cells at a distance from the wound edge. To do this, I prepared extracts from cells

>3 mm from the site of agarose strip removal (for details of the procedure, see Chapter One,

Figure 6). Inclusion of apyrase in the medium had no detectable effect on basal levels of EGFR

activity, but it abolished wound-induced stimulation (Figure 11). Similarly, wound-induced ERK

activation was abrogated by treatment with apyrase. Although I did not measure ATP

concentrations following agarose strip removal, the presence of apyrase-sensitive signals

indicates that ATP is released from cells in this wounding model. As a control, I noted that

treatment with apyrase did not reduce the activation of EGFR and ERK1/2 kinases after addition

of EGF (Figure 12).

24
 Figure 11 Extracellular ATP is necessary for EGFR activation near the

wound edge

Western Blots for activation of EGFR and ERK1/2 after no treatment (NT),

wounding (W), and/or treatment with 30 U/ml Apyrase (APY). * indicates

significant reduction from wounded controls (p<0.001).

Figure 12 EGFR and ERK1/2 activation by EGF is not inhibited by

apyrase or PLD2 siRNA

Western Blots for activation of EGFR and ERK1/2 after no treatment

(NT), 10 min incubation with 10 ng/ml EGF, and/or treatment with 30

U/ml Apyrase (APY) or transfection with 50 nM PLD2 siRNA (see

Chapter Three).

To visualize the extent of EGFR signaling induced by extracellular ATP directly, I

performed immunofluorescence microscopy following stimulation by removal of an agarose

strip. I was not able to identify an antibody to the activated EGFR that generated a satisfactory

read-out, but phospho-ERK1/2 antibodies produced good signals. ERK1/2 were activated at least

up to 5 mm from the wound edge (Figure 13). The specificity of the immunofluorescence

signals was verified by the fact that UO126, which inhibits ERK1/2 activation, quenched the

signals, and the EGFR-dependence of the signals was verified by the observation that they were

blocked by the EGFR tyrosine kinase inhibitor AG 1478. Inclusion of apyrase in the medium

resulted in quenching of the signals at a distance from the wounds, but ERK1/2 activation was

still clearly detected within 250 m of the wound edge. Together, these results indicate that

25
extracellular ATP is necessary for EGFR activation >250 m from the wound edge, but that

extracellular ATP-independent mechanisms function in cells near the wound edge.

Figure 13 Activation of the EGFR/ERK1/2 pathway is dependent on signaling by ATP in cells far from, but

not near wounds

Sheets of HCLE cells were cultured around an agarose strip and were left unwounded (No Wound), wounded

(Wound), or wounded in the presence of 30 U/ml apyrase (Wound +APY), fixed 10 minutes later, and stained with

antibodies recognizing the activated forms of ERK1/2. Results of quantitative analysis are shown below the

immunofluorescence images, in the same scale. Images of cells wounded in the presence of 10 M UO126 (Wound

+ UO) or 1 M tyrphostin AG 1478 (Wound +AG) were also analyzed.

26
2.2.3 Wounds heal in the absence of signaling by extracellular ATP.

Cells within monolayers typically move spontaneously, and according to some models, cells do

not move into wounds as a result of increases in speeds of migration, but rather as a result of the

lack of mechanical constraints in the denuded area (Bindschadler and McGrath, 2007; Sherratt

and Dallon, 2002). According to such models, it is not necessary to postulate the existence of any

biochemical signals that affect cell behavior. To determine whether wounding induces increased

velocities, positions of cells were recorded at various times by time-lapse microscopy. As is seen

in Figure 14, cells within 50 m from the agarose barrier exhibited basal motility irrespective of

induction of wounds. However, wounding induced significant increases of the velocities of the

cells at the edge of wounds, and this was not affected by the presence of apyrase. I conclude that

wounding, by simply removing a physical barrier, does indeed elicit a biological response in the

HCLE cells in that they move faster.

Figure 14 Increased velocities of cells near the wound

edge are not dependent on extracellular ATP.

Agarose strips were left in place (NT) or removed

(Wound) in the presence of 30 U/ml apyrase (APY), as

indicated. The velocities of individual cells within 50 m

of the wound edge (Edge cells) or greater than 2 mm from

the wound edge (Distal cells) were calculated. * indicates

significant increases from unwounded groups (p<0.001).

I next tested whether signaling by ATP is necessary for healing of wounds by examining

cell migration in the presence of 30 U/ml apyrase, which reduced extracellular ATP to

27
undetectable levels (Figure 9) and inhibited wound-induced PKC activation (Figure 10A). As

shown in Figure 15A, no significant difference in wound healing was observed between the

untreated control and apyrase-treated groups. Similar results were obtained with secondary

cultures of rabbit corneal epithelial cells (Figure 15B) and with cultured rabbit corneas (Figure

15C).

Figure 15 Extracellular ATP is not

necessary for wound healing

Healing of A. HCLE cells, B. secondary

cultures of rabbit corneal epithelial

cells, and C. cultured rabbit corneas

were monitored in the presence of 30

U/ml apyrase (APY), as indicated. No significant differences were observed.

To verify that the apyrase was functional, I tested its ability to degrade [-32P]-ATP and

[-32P]-ADP. As little as 0.3 U/ml apyrase was sufficient to degrade ATP and ADP completely,

and 30 U/ml was fully active even after 14 hours in cell culture (Figure 16). From these data, I

conclude that extracellular ATP signaling is not necessary for wound healing.

Figure 16 Apyrase was active throughout the healing

assays

Wounded sheets of HCLE cells were incubated with the

indicated concentrations of apyrase and the degradation of

[-32P]-ATP was measured for 5 min. The 5-minute

degradation of ATP was also evaluated in the medium of

cells cultured for 14 hours with 30 U/ml apyrase (*30*).

28
 Other reports have suggested that ATP signaling is critical for wound healing because

healing and wound-induced EGFR activation are inhibited by the general P2 receptor antagonist

reactive blue 2 (RB2) (Boucher et al., 2007; Klepeis et al., 2004; Yin et al., 2007). I have

confirmed these findings but I have also observed that the same concentration of RB2 inhibited

activation of the EGFR by AR, which makes RB2 an unsuitable reagent for this type of study

(Figure 17).

Figure 17 RB2 inhibits healing and EGFR

activation by AR.

A. Healing of wounds in HCLE cells with no

treatment (NT) or 100 M RB2. B. Cells

received no treatment (NT) or 10 ng/ml AR for

10 minutes (AR), following 30 minute pre-

treatment with 100 M RB2 as indicated.

2.3 DISCUSSION

Previous reports have shown that wounding sheets of corneal epithelial cells results in release of

ATP, and that extracellular ATP can activate the EGFR (Boucher et al., 2007; Yin et al., 2007).

The data presented in this section indicate that ATP is necessary for wound-induced EGFR and

ERK1/2 activations in cells far (>250 m) from wounds. Additionally, I report here the

existence of a distinct mode of activation of the EGFR after wounding, which is not dependent

on ATP signaling, based on the following observations: 1, activation of the EGFR in cells near

the wound edge is unaffected by removal of extracellular ATP with apyrase; 2, visualization by

29
immunofluorescence shows that the EGFR/ERK1/2 pathway is stimulated in cells near wounds

in the presence of apyrase; 3, the enhanced velocity of cells near wounds is not affected by the

presence of apyrase.

Elimination of extracellular ATP with apyrase did not affect the rate of wound healing,

suggesting that extracellular ATP signaling pathways are not required for healing of wounds.

This is in agreement with one previous report describing similar results with Madin-Darby

canine kidney epithelial cells (Farooqui and Fenteany, 2005), but differs from the conclusions

reached by the use of the purinergic receptor antagonist RB2 (Boucher et al., 2007; Klepeis et al.,

2004; Yin et al., 2007). Importantly, I have found that RB2 also blocks EGFR activation by

exogenous ligands. RB2 is therefore not a suitable reagent for determining the role of purinergic

signaling in healing of wounds in corneal epithelial cells because EGFR activation is an absolute

prerequisite for the induction of motility. Numerous P2Y receptors have been found in the

corneal epithelium (Klepeis et al., 2004), but to date, no knockout studies have addressed

whether these receptors regulate wound healing in vivo.

Differences in signaling based on distance from the wound edge have been described

previously. For instance, in human corneal epithelial cells immortalized by the SV-40 T-antigen,

increased cytosolic [Ca2+] in edge cells is dependent on extracellular calcium but the propagation

of a calcium wave to distal cells is dependent on intracellular calcium (Klepeis et al., 2001).

Also, in Madin-Darby canine kidney epithelial cells, the presence of reactive oxygen species and

the activation of ERK1/2 was observed only in near-edge cells 10 minutes after wounding

(Nikolic et al., 2006). Methods that allow analysis of signaling and physiology selectively in

either wound-proximal or distal cells will continue to help define the functions of these distinct

cell populations in wound healing.

30
 A notable difference in the two EGFR activation pathways is their range of action. ATP

is diffusible, and I found both by direct immunoblotting of cells at a distance from wounds and

by immunofluorescence studies that ATP signaling activates the EGFR up to at least 5 mm from

the wound edge. ATP has previously been shown to mediate intracellular [Ca2+] increases over

similar distances in mechanically stimulated liver and alveolar epithelial cells and wounded

corneal epithelial cells (Frame and de Feijter, 1997; Isakson et al., 2001; Klepeis et al., 2001).

Also, ATP is found in conditioned media after wounding at a concentration of 1-2 M, which is

sufficient to activate the EGFR (Yin et al., 2007).

In contrast to ATP signaling, the extracellular ATP-independent signaling pathway

appears to act only in cells near wounds. EGFR ligands such as AR and HB-EGF are not

expected to mediate strong activation of the EGFR in wound-distal cells because of their limited

diffusion. HB-EGF and AR bind to not only the EGFR but also to cell surface

glycosaminoglycans and the tetraspannin CD9, resulting in an active ligand that remains mostly

cell-associated (Harris et al., 2003; Iwamoto et al., 1994). I detected AR in wound-conditioned

media at concentrations below that which induces detectable activation of the EGFR (<100

pg/ml), yet the EGFR was highly activated after wounding. However, even when EGFR

activation cannot be detected by immunoblotting, the EGFR may still transduce potent biological

effects, especially when the ligand can limit receptor internalization (Iyer et al., 2008; Reddy et

al., 1996). Thus, after wounding, AR acts predominantly in a local manner while ATP signals

over distances.

31
2.3.1 Summary and Model

In summary, wounding activates the EGFR by dual mechanisms: by extracellular ATP signaling

and by an unknown mechanism near the wound edge (see diagram in Figure 18). The

mechanism by which ATP transactivates the EGFR is explored in Chapter Three, and

mechanisms of ATP release from wound edge cells are discussed in Chapter Four. Mechanisms

of extracellular ATP-independent EGFR activation near the wound edge are the subjects of

Chapters Five and Six.

Figure 18 A hypothetical model for the role of extracellular ATP in wound-induced EGFR activation

See text for details. P2R, type-2 purinergic receptor; MMP, matrix metalloprotease; ADAM, a disintegrin and

metalloprotease; EGFR, epidermal growth factor receptor; ERK1/2, extracellular signal-regulated kinases 1 and 2.

32
 3.0 ATP TRANSACTIVATES THE EGFR THROUGH PHOSPHOLIPASE D2

SIGNALING

3.1 INTRODUCTION

In Chapter Two, I confirmed that extracellular ATP transactivates the EGFR after wounding.

Similarly to signaling by extracellular ATP, signaling by Phospholipase D (PLD) has recently

been identified as a potential regulator of wound-induced EGFR activation (Mazie et al., 2006).

PLD catalyzes the hydrolysis of phosphatidylcholine, an abundant membrane phospholipid, to

free choline and the second-messenger phosphatidic acid (PA) (Cazzolli et al., 2006; Exton,

2002; Jenkins and Frohman, 2005; McDermott et al., 2004). PLD is activated rapidly after

wounding sheets of corneal epithelial cells and treatment with PA activates the EGFR (Mazie et

al., 2006). In fact, the EGFR transactivation processes induced by exogenously added ATP and

PA both depend on proteolytic ligand release, and ATP is known to induce activation of

phospholipase D (PLD) in many systems (el-Moatassim and Dubyak, 1992; Gargett et al., 1996;

Kusner and Adams, 2000; Perez-Andres et al., 2002; Pochet et al., 2003; Sun et al., 1999). This

led me to test the hypothesis that ATP signals through PLD to activate the EGFR.

The observation that healing rates of HCLE and rabbit corneal epithelial cells (Figure 15)

and of Madin Darby canine kidney epithelial cells (Farooqui and Fenteany, 2005) are not

affected by treatment with apyrase suggests that EGFR transactivation by extracellular ATP has

33
a biological role in addition to enhancing cell motility. Although wounds heal due to cell

migration, a normal response to wounding includes replacing lost cells via cell division, which is

also regulated by EGFR activity (Nakamura et al., 2001). In the corneal epithelium, wounding

induces cell division in the limbus (see Figure 1) (Barrandon, 2007; Daniels et al., 2006;

Revoltella et al., 2007; Zieske et al., 2001), an anatomical region that is typically distal from the

point of injury. The observation that extracellular ATP stimulates cell division in skin

keratinocytes (Braun et al., 2006) therefore suggests that EGFR transactivation by ATP may be a

wound-induced mitogenic signal. In this chapter, I have examined whether PLD mediates EGFR

transactivation by extracellular ATP and whether extracellular ATP promotes cell division after

wounding. Portions of this chapter were published previously, ASCB copyright 2008, in

http://www.molbiolcell.org/cgi/content/full/19/11/4909 (Block and Klarlund, 2008) and have, in

some cases, been modified.

3.2 RESULTS

3.2.1 ATP is released after wounding and activates PLD.

I have previously shown that wounding sheets of HCLE cells by the removal of agarose droplets

results in the release of ATP (Figure 9) and in activation of PLD (Mazie et al., 2006). To

determine whether PLD is activated as a consequence of the released ATP, cell sheets were

wounded in the presence of apyrase. Apyrase eliminated ATP effectively from the medium

(Figure 9), and apyrase reduced wound-induced PLD activity to unwounded levels (Figure 19),

which implies that activation of PLD is stimulated by the ATP that is released after wounding.

34
 3.1

Figure 19 ATP released after wounding activates PLD

Assay for activity of PLD after no treatment (NT), wounding (W),

and/or treatment with 30 U/ml apyrase (APY) and incubation for 5

minutes. * indicates significant reduction from wounded controls

(p<0.001).

3.2.2 The PLD2 isoform is activated by wounding and by treatment with ATP.

No satisfactory chemical inhibitor of PLD activity was commercially available so I employed a

short interfering RNA (siRNA)-based approach to evaluate whether PLD is necessary for ATP-

stimulated EGFR activation. Mammalian cells express two PLD isoforms, PLD1 and PLD2,

which are both ubiquitously expressed. Distinct intracellular localizations, mechanisms of

activation, and downstream effectors have been described for each isoform (Hiroyama and

Exton, 2005; Kusner et al., 2002; Morris, 2007; Oude Weernink et al., 2007; Preininger et al.,

2006). SiRNA oligonucleotides targeted to PLD1 and PLD2 were transfected into HCLE cells

and whole cell extracts were prepared 3 days after transfection, and as illustrated by the

immunoblots in Figure 20A, the siRNA-mediated knock-downs of the PLD1 or PLD2 isoforms

were efficient and specific. Quantitation of the intensities of the bands shows that expression of

PLD2 was reduced by 80% and PLD1 by 70% (Figure 20B). PLD activation after wounding was

greatly diminished in cells transfected with one of the PLD2 oligonucleotides (PLD2 siRNA-a)

but was unaffected in cells transfected with PLD1 siRNA, and knock-down of both isoforms had

no greater inhibitory effect than knock-down of the PLD2 isoform alone (Figure 20C). A second

PLD2 siRNA oligonucleotide (PLD2 siRNA-b), which was made to control for off-target effects,

also reduced the level of PLD2 expression. These results suggest that wound-induced PLD

35
activation is predominantly determined by the PLD2 isoform. In the following experiments with

siRNAs, PLD1 and 2 were verified to be downregulated to levels similar to that shown in Figure

20A.

Figure 20 Wounding selectively activates the PLD2 isoform

A. Western blots of whole cell extracts of HCLE cells untransfected (No siRNA) or transfected with siRNA against

PLD1 or PLD2 cultured around agarose droplets. B. Densitometry of immunoblots transfected with the indicated

siRNAs. C. Assay for activity of PLD after no treatment (NT), wounding (W), and/or transfection with siRNAs

against PLD1 or PLD2 . * indicates significant reduction from wounded controls (p<0.001).

Exogenously added ATP stimulated PLD activity in HCLE cells (Figure 21A), as it does

in many other cell types (el-Moatassim and Dubyak, 1992; Gargett et al., 1996; Kusner and

Adams, 2000; Perez-Andres et al., 2002; Pochet et al., 2003; Sun et al., 1999). Similarly to

wound-stimulated PLD activity, ATP-stimulated PLD activity was blocked in HCLE cells

transfected with PLD2 siRNA-a but not in cells transfected with PLD1 siRNA (Figure 21A).

Stimulation with apyrase-treated ATP or with 50 M adenosine did not activate PLD (Figure

21B), indicating that activation is a direct effect of ATP. Addition of 10 M of the EGFR kinase

inhibitor tyrphostin AG 1478 did not inhibit activation of PLD by ATP (Figure 21B) although it

36
completely inhibited EGFR activation by ATP in parallel incubations, indicating that the

observed PLD activation is not a result of EGFR signaling.

Figure 21 Extracellular ATP

activates PLD2

A. PLD activities in confluent HCLE

cells not treated (NT) or transfected

with siRNA against PLD1, PLD2, or

both and then treated for 5 minutes

with 50 M ATP. * indicates

significant reduction from ATP-

treated controls (p<0.001). B. Assay for activity of PLD after no treatment (NT), 5 minute incubation with 50 M

ATP (ATP), 50 M ATP that had been pre-incubated for 15 minutes with 5 U/ml apyrase (ATP+APY), or 50 M

ATP after 15 minute pre-treatment of cells with 10 M tyrphostin AG 1478 (ATP+AG).

3.2.3 Extracellular ATP transactivates the EGFR via PLD2 signaling.

To test the role of PLD signaling in ATP-stimulated EGFR activation, cells were transfected

with siRNA oligonucleotides. Activation was monitored by Western Blotting with an antibody

that recognizes the EGFR phosphorylated on tyrosine 1173. Both PLD2 siRNA-a and siRNA-b

blocked EGFR activation, whereas PLD1 siRNA did not (Figure 22A and B). Transfection of

PLD2 siRNA-a did not block activation of the EGFR by EGF (Figure 12). This provides direct

evidence that activation of the EGFR by ATP is mediated through PLD2 signaling.

37
 Figure 22 ATP activates the

EGFR through PLD2

signaling

Western Blotting and

densitometry for EGFR

activation after no treatment

(NT) or treatment with 50 M

ATP for 10 minutes (ATP). Cells were transfected with siRNA against A. PLD1 or PLD2 (siPLD2a) or B. a

different PLD2 siRNA (siPLD2b). * indicates significant reduction from ATP-treated controls (p<0.001).

Extracellular ATP transactivates the EGFR through stimulation of proteases at the cell

surface that cleave precursors of ligands for the receptor (Boucher et al., 2007; Yin et al., 2007).

To establish a measure of the transactivation process, I assayed one such ligand, amphiregulin

(AR), and found that wounding, addition of ATP, and stimulation with a water-soluble analog of

phosphatidic acid 1,2-octanoyl-sn-glycero-3-phosphate (C8-PA), all increased AR release (Figure

23). Furthermore, neutralization experiments indicated that AR contributes to EGFR activation

Figure 23 AR is released after wounding or treatment with PA or ATP

Quantities of AR in conditioned media were determined in the 10 minutes following no treatment (NT) or A.

wounding, B. treatment with 25 M C8PA (PA), C. treatment with 50 M ATP, and or 30 U/ml apyrase. * indicates

significant increase from untreated controls (p<0.001).

38
in response to these treatments (Figure 24).

Figure 24 AR contributes to EGFR activation after wounding or treatment with PA or ATP

Activation of the EGFR was determined by densitometry of Western Blots. Cells were stimulated for 10 minutes

after A. wounding, B. treatment with 25 M C8PA (PA), or C. treatment with 50 M ATP. As indicated, cells were

incubated for 2 hr with 20 g/ml neutralizing antibodies against TGF (isotype control) or AR. * indicates

significant decrease from stimulated controls (p<0.001).

Because ATP activates the EGFR through PLD2, reduction of expression of PLD2 was

expected to decrease secretion of AR in response to ATP. Reducing the level of PLD2 resulted

in reduction of basal and ATP-stimulated secretion of AR, whereas interfering with the cellular

levels of PLD1 had no effect (Figure 25A). The reduction of PLD2 by either siRNA

oligonucleotide correlated directly with AR release, suggesting that the reduction of AR release

is the result of knock-down of PLD2 rather than of off-target effects (Figure 25B).

39
 Figure 25 ATP-

stimulated AR release

depends on PLD2

A. Quantities of AR in

conditioned media were

determined in the 10

minutes following no

treatment (NT) or

treatment with 50 M ATP. As indicated, cells were transfected with siRNA against PLD1 or PLD2. * indicates

significant decrease from untreated controls and ** indicates significant decrease from ATP-treated controls

(p<0.001). B. HCLE cells were untransfected (data points farthest to the right) or transfected with PLD2 siRNAa or

b at 50, 16, 5.0, or 1.6 nM (data points from left to right). Cells were then left untreated (NT) or were treated with

50 M ATP before determination of AR concentration in conditioned medium and cellular PLD2 expression.

3.2.4 PLD2 activation is not required for wound-proximal EGFR activation

Together with results from Chapter Two, data presented in this section suggest that ATP/PLD2

signaling is necessary for EGFR transactivation in cells >250 m from the wound edge. To

determine if PLD2 is necessary for signaling near the edge, I examined EGFR activation in cells

transfected with PLD2 siRNA. As shown in Figure 26, knockdown of PLD2 did not inhibit

EGFR activation near the wound edge, confirming my observations from Chapter Two (Figure

11) that EGFR activation in wound-edge cells proceeds independently of extracellular ATP

signaling.

40
 Figure 26 EGFR activation near the edge is independent of

signaling by PLD

Western Blots for wound-induced activation of EGFR in the

presence of no treatment (NT) or transfection with siRNAs against

PLD1 and/or 2.

3.2.5 The mechanism of PLD activation by extracellular ATP

The mechanism of PLD activation by extracellular ATP is not entirely clear. PLD activity

requires binding to phosphatidylinositol 4,5-bisphosphate and activity of protein kinase C (PKC)

(Exton, 2002; Jenkins and Frohman, 2005; McDermott et al., 2004). Extracellular ATP binds to

the P2X class of cation channels and the P2Y class of G-protein coupled receptors (Abbracchio

et al., 2006). ATP may also be metabolized by ecto-enzymes to adenosine, which activates the

P1 class of receptors. Since apyrase abrogated activation of PLD after wounding and

exogenously-added adenosine had no effect on PLD activity, P1 receptors are not likely involved

in PLD2 activation. Agonist and antagonist profiling has indicated that PLD is activated through

P2X7 and P2Y2 receptors that signal through phospholipase C (PLC) and protein kinase C (PKC)

(el-Moatassim and Dubyak, 1992; Katzur et al., 1999; Purkiss et al., 1992). In preliminary

studies, I have observed that wound-induced PLD activation and ATP-stimulated AR release are

both sensitive to the classical PKC inhibitor bisindolylmaleimide I (Figure 27), and that

wounding activates PKC through extracellular ATP signaling (Figure 10), which is consistent

with the hypothesis that extracellular ATP activates PLD through PLC and PKC signaling.

41
 Figure 27 PKC activity is

necessary for wound-induced

PLD activation and ATP-induced

AR release

Assays for A. PLD activity or B.

AR release following no treatment (NT), wounding (W), or treatment with 50 M ATP. 10 M of the Classical

PKC inhibitor bisindolylmaleimide I (Bis) was added 30 minutes prior to stimulations. * indicates significant

decrease from ATP-treated controls (p<0.001).

3.2.6 PLD2 mediates EGFR activation through SFKs

Recent research has indicated that EGFR transactivation by GPCR ligands (such as ATP) is

dependent on activity of the Src-family kinases (SFKs) (Zhang et al., 2004). To determine

whether SFK activity is necessary for EGFR transactivation by extracellular ATP in the present

system, I incubated cells with PP2 or Src Kinase Inhibitor-I (SKI), two structurally dissimilar

SFK inhibitors that are known to have distinct non-specific interactions (Bain et al., 2007).

Either SFK inhibitor blocked EGFR activation by treatment with ATP (Figure 28A) but not by

treatment with EGF (Figure 28B). Similarly, the SFK inhibitors blocked ATP-stimulated AR

release (Figure 28C). These results suggest that ATP requires SFK activity to transactivate the

EGFR. I next determined whether SFK signaling acts upstream of PLD2. The SFK inhibitors

did not reduce PLD activation by wounding (Figure 29A) or by treatment with ATP (Figure

29B). Together, these results indicate that SFK activity acts downstream of PLD2 to mediate

EGFR transactivation by ATP.

42
Figure 28 Activation of SFKs is necessary for EGFR transactivation by extracellular ATP

A. Western Blotting and densitometry for activation of the EGFR after no treatment (NT) or 10 minute treatment

with 50 M ATP. B. Cells were treated as in A but received 10 ng/ml EGF instead of ATP. C. AR quantities were

determined in conditioned media from cells treated as in A. * indicates significant decrease from ATP-treated

controls (p<0.001).

Figure 29 Wound- and ATP-induced activation of PLD proceeds independently of SFK activation

Assay for PLD activity following no treatment (NT), A. wounding (W), or B. treatment with 50 M ATP. 10 M

PP2 or was added 30 minutes prior to stimulations. Wound- or ATP-stimulated groups were significantly increased

from unstimulated controls but were not significantly affected by PP2.

PLD can affect signaling through intermolecular interactions (Kim et al., 2006; Lee et al.,

2006) or through the direct actions of PA (Fang et al., 2001; Frank et al., 1999; Lehman et al.,

2007; Zhao et al., 2007). Since PA stimulated AR release and EGFR activation (Figures 23 and

43
24), PLD2 signaling likely stimulates EGFR activation through the actions of PA. Treatment of

cells with PA stimulated activation of SFKs, as indicated by Western Blotting for the kinases

autophosphorylated on Tyr-419 (Figure 30A), and SFK inhibitors blocked EGFR activation by

treatment with PA (Figure 30B). These results are consistent with the hypothesis that after

wounding, ATP transactivates the EGFR through the sequential activation of PLD2 and SFKs.

Figure 30 PA activates the EGFR through

SFK signaling

Western Blotting for activation of A. SFKs or

B. the EGFR following no treatment (NT) or

treatment with 25 M C8PA (PA). 10 M PP2

was added 30 minutes prior to stimulation.

3.2.7 Wound-induced cell division depends on signaling by EGFR but not extracellular

ATP.

Transactivation of the EGFR by extracellular ATP can promote cell motility, but it is unclear

whether this signaling module serves any additional biological role after wounding. EGFR

activation can stimulate cell division in many cell types, and regulates wound-induced

proliferation in the corneal epithelium (Nakamura et al., 2001). Furthermore, PLD signaling is

mitogenic (Foster and Xu, 2003) and in cultured keratinocytes, extracellular ATP promoted cell

division (Braun et al., 2006). Therefore, I investigated whether extracellular ATP regulates cell

division after wounding. To detect cell division, I immunostained for Ki67, a protein present in

the nuclei of dividing cells (Brown et al., 2002).

To determine whether EGFR transactivation by ATP is necessary for cell division, I

examined Ki67 immunostaining after wounding sheets of HCLE cells. Wounding caused

44
increased Ki67 staining in cells near the wound margin at 8 and 24 hours, and was most

increased at 48 hours, when the wounds had fully healed (Figure 31). Wounding did not alter the

number of Ki67-positive cells in areas >1 mm (~50 rows of cells back) from the wound edge,

which was similar to that observed near the edges of unwounded controls. Treatment with the

EGFR kinase inhibitor AG 1478 reduced Ki67 staining, but treatment with apyrase had no effect

on either the number or location of Ki67-positive cells (Figure 31). These results indicate that

cell division stimulated by wounding is regulated by activation of the EGFR, independently of

signaling by extracellular ATP.

Figure 31 Wound-induced cell division depends on signaling by the EGFR but not extracellular ATP

Immunofluorescence images taken from HCLE cell cultures 48 hours after wounding. Cells were stained for E-

cadherin (green) and Ki67 (red). As indicated, cells were treated with 1 M of the EGFR inhibitor AG 1478 (AG)

or 25 U/ml apyrase (Apy) during healing. The wound edge is visible in the No Wound and Wound + AG

conditions.

45
 3.3 DISCUSSION

PLD activation and release of extracellular ATP have both been suggested to be up-stream

events that lead to EGFR transactivation after wounding (Boucher et al., 2007; Klepeis et al.,

2001; Klepeis et al., 2004; Mazie et al., 2006; Yin et al., 2007). In this chapter I have reported

that these two signaling events are related by showing that ATP signals through PLD2 to activate

the EGFR. I have based this conclusion on the following evidence: 1, wound-induced PLD

activation is abrogated by apyrase; 2, wounding and treatment with ATP both activate the PLD2

isoform; 3, treatment with ATP or PA similarly stimulates EGFR activation through AR and

SFK signaling; and 4, knock-down of PLD2 expression inhibits ATP-induced EGFR activation

and AR release.

ATP/PLD2 signaling is necessary for EGFR activation in cells >250 m from the wound.

This pathway contributes to the induction of motility since addition of extracellular ATP or PA

has been shown to enhance wound healing in many different epithelial cell lines (Dignass et al.,

1998; Klepeis et al., 2004; Mazie et al., 2006; Sponsel et al., 1995; Wesley et al., 2007; Yin et

al., 2007). I was unable to use siRNA to test whether the activity of PLD2 is necessary for

wound healing. Only stratified cultures of HCLE cells heal efficiently and PLD2 expression in

siRNA-transfected cells was restored to control levels after the stratification process. An

alternate approach to examine the effects of PLD signaling on cell migration would be to use a

class of recently-described chemical inhibitors that were not available at the time of this research

(Lavieri et al., 2009; Lewis et al., 2009). Traditionally, primary alcohols such as 1-butanol have

been used to inhibit PLD activities, despite numerous non-specific and toxic effects.

Signaling by the EGFR, PLD, and extracellular ATP are known to stimulate cell division

(Braun et al., 2006; Foster and Xu, 2003; Nakamura et al., 2001). However, apyrase had no

46
effect on wound-induced cell division (Figure 31), indicating that extracellular ATP is not

necessary for cell division after wounding. In vivo, wounding stimulates corneal epithelial cell

division within the wound margin (Zieske et al., 2001), as I observed in Figure 31, and far from

the wound in the limbus, which is the niche for epithelial stem and transit amplifying cells

(Daniels et al., 2006). I attempted to investigate whether extracellular ATP regulates cell

division in the limbus, but I was unable to reproduce the phenomenon of wound-induced limbal

cell division in mouse and rat whole eye organ cultures. In the future, one could treat mouse

eyes with apyrase in vivo or use a gene knockout approach to test the hypothesis that ATP/PLD2

signaling regulates cell division in the limbus. Potential mechanisms of ATP release and

additional biological outcomes of extracellular ATP signaling are discussed in Chapter Four.

My observation that treatment with PA induces SFK activation (Figure 30) is in

agreement with the observation that PLD2 over-expression increased SFK activity (Ahn et al.,

2003). However, the mechanism whereby PA activates SFKs is unknown. SFK activity is

stimulated by intermolecular interactions and de-phosphorylation of the Tyr-527 residue. A

number of candidate phosphatases have been identified as mediating SFK activation by de-

phosphorylating Tyr-527 including PTP1B, Shp1 (Src homology 2 domain-containing tyrosine

phosphatase 1), and Shp2 (Roskoski, 2005). Interestingly, PA has been shown to bind and

promote the activity of Shp1 (Frank et al., 1999). The hypothesis that PA activates SFK through

the activation of Shp1 could, in the future, be tested by measuring Shp1 activity in response to

PA and by inhibiting Shp1 activity with siRNA.

The results reported here are to my knowledge the first descriptions of a role for PLD in

ATP-stimulated EGFR transactivation. There is a growing list of cytokines and growth factors

that stimulate EGFR transactivation (Higashiyama and Nanba, 2005; Ohtsu et al., 2006;

47
Sanderson et al., 2006) and many of these, such as angiotensin II, bradykinin, endothelin-I, and

lysophosphatidic acid also stimulate PLD activity (Bollag et al., 1990; Liu et al., 1992; Martin et

al., 1989; van der Bend et al., 1992). A role for PLD2 has previously been reported for

angiotensin II-mediated transactivation of the EGFR (Li and Malik, 2005), and I have also

observed that AR release stimulated by phorbol 12-myristate 13-acetate (PMA or TPA) is

dependent on PLD2. It therefore seems reasonable to hypothesize that PLD2 mediates EGFR

transactivation by stimuli other than ATP.

3.3.1 Summary and model

In summary, ATP is released from cell sheets after wounding by the removal of agarose barriers.

Extracellular ATP activates the PLD2 isoform leading to the production of PA, which in turn

activates SFK, thereby stimulating ligand release and EGFR activation (see diagram in Figure

32). EGFR activation, but not extracellular ATP, is necessary for wound-induced cell division.

48
Figure 32 A hypothetical model for the mechanism of EGFR trans-activation by extracellular ATP after

epithelial wounding

See text for details. P2R, type-2 purinergic receptor; PLD2, phospholipase D2; PA, phosphatidic acid; SFK, Src-

family kinase; MMP, matrix metalloprotease; ADAM, a disintegrin and metalloprotease; EGFR, epidermal growth

factor receptor.

49
 4.0 ATP IS RELEASED PERSISTENTLY FROM WOUNDS AND FREE EDGES IN

EPITHELIAL SHEETS

4.1 INTRODUCTION

Data presented in Chapters Two and Three and from other laboratories show that ATP is

released from wounded epithelia and transactivates the EGFR (Block and Klarlund, 2008;

Boucher et al., 2007; Yin et al., 2007). I have not observed a necessity for extracellular ATP in

wound healing or cell division. However, extracellular ATP can promote epithelial cell motility

(Dignass et al., 1998; Klepeis et al., 2004; Sponsel et al., 1995; Wesley et al., 2007; Yin et al.,

2007) and may additionally have effects on wound healing and ocular physiology that cannot be

observed in cultured epithelial cells. ATP is now recognized as an extracellular messenger with

numerous functions, and has the potential of influencing processes related to wound healing such

as induction of inflammation, regeneration of nerves, alteration of tear composition, and

communication with the underlying stroma (Belliveau et al., 2006; Bours et al., 2006;

Burnstock, 2006; Burnstock, 2007; Chen et al., 2006; Jumblatt and Jumblatt, 1998). It is

therefore important to understand the kinetics and mechanisms of ATP release from wounded

sheets of epithelial cells.

The concentration of ATP is roughly 1000-fold higher inside of cells, so there is a large

outwardly directed gradient. Consequently, ATP is released into the extracellular space from

50
ruptured cells after wounding. Additionally, wounding models that minimize cell damage have

indicated that ATP may be released after wounding independently of cell rupture (Figures 9 and

11). ABC-family transporters such as P-glycoprotein and CFTR, fusion and release of ATP-

loaded vesicles, pannexin channels, and gap junction hemichannels have all been described as

conduits for ATP release (Lazarowski et al., 2003; Praetorius and Leipziger, 2009; Schwiebert

and Zsembery, 2003; Spray et al., 2006).

Gap junctions are cell surface structures that allow the transfer of small molecules (less

than 1000 Da) between cells (Evans and Martin, 2002). Each cell expresses half of the channel,

a hemichannel, which is itself a hexamer of proteins of the connexin family. Unpaired

hemichannels, or connexons, are expressed at the cell surface and secrete small molecules such

as ATP in regulatable fashion (Evans et al., 2006; Goodenough and Paul, 2003; Li et al., 1996).

The corneal epithelium expresses functional gap junctions that are composed of the connexins

(named according to molecular weight) Cx43, Cx31.1, Cx30, and Cx26 (Chen et al., 2006;

Shurman et al., 2005; Williams and Watsky, 2002). Evidence for ATP release through

hemichannels includes a number of studies employing both gain-of-function (connexin

expression) and loss-of-function (connexin inhibition) methodologies in HeLa and other cell

lines (Cotrina et al., 1998), astrocytes (Stout et al., 2002), corneal endothelial cells (Gomes et al.,

2005), cochlear-supporting cells (Zhao et al., 2005), and osteocytes (Genetos et al., 2007).

Since I have detected ATP release from wounded cell cultures containing mostly wound-

proximal cells ((Block and Klarlund, 2008) and Figure 9), I expect that ATP is released from

cells near the wound edge. Here, I used multiple wounding models to examine the kinetics of

ATP release from cells near the wound edge, and I correlated this with markers of cell rupture. I

51
have also conducted preliminary experiments to test the hypothesis that ATP is released from

cells near the wound edge through connexin hemichannels.

4.2 RESULTS

4.2.1 ATP is released from cells near edges in epithelial sheets chronically and

independently of damage.

Using wounding models that minimize cell damage, I have observed that ATP is released acutely

after wounding epithelial sheets ((Block and Klarlund, 2008), Figures 9 and 11), but even

minimal cell damage can lead to ATP leaking from cells. Cells respond to sublethal damage by a

variety of repair mechanisms that re-seal the plasma membrane (Bement et al., 2007; Idone et al.,

2008; McNeil and Kirchhausen, 2005; Woolley and Martin, 2000), but until repair is complete,

lactate dehydrogenase (LDH) leaks out of cells (Loo and Rillema, 1998). Increased amounts of

LDH were detected in culture supernatants in the ten minutes immediately following the removal

of agarose droplets (Figure 33A). For comparison, I note that levels of LDH released following

agarose removal were roughly 10% of those observed following scrape-wounding. Cells were

washed and fresh conditioned media were collected at various later time points. Increased LDH

was detected after 20-30 minutes, but was no different from unwounded controls after one or 24

hours (Figure 33A). The same conditioned media were analyzed to determine quantities of

extracellular ATP, which can be measured with a bioluminescence assay (see Methods 8.19). As

shown in Figure 33B, elevated ATP concentrations were observed at all time points. Together,

52
these data indicate that cells near wound edges secrete ATP persistently and independently of

cell damage.

Figure 33 ATP is released persistently after wounding.

Assays for quantities of A. LDH or B. ATP in conditioned media from HCLE cell sheets collected at the indicated

times after wounding by removal of agarose droplets.

Although I could not detect increased LDH in the supernatants of cultures 24 hours after

wounding, these cells had experienced some degree of rupture initially. Therefore, to verify that

cells near wound edges secrete ATP independently of cell damage, I employed a different model

for analyzing cells near edges in epithelial sheets. To produce a cell culture with many free

edges, HCLE cells were cultured on thin plastic strips above a non-adhesive surface of

polyHEMA (see Figure 7 and Methods 8.6). As controls, cells were cultured on polyHEMA-

coated dishes that had been completely covered with plastic. Levels of lactate dehydrogenase in

the supernatants of cells grown in the two conditions were low and similar, indicating that

growth of cells on the strips does not promote cell lysis (Figure 34A). Importantly, ATP

concentrations were significantly higher in supernatants from cells grown on strips (Figure 34B).

These results corroborate the data gathered from wounded cell sheets indicating that cells near

the wound edge secrete ATP persistently and in the absence of cell damage.

53
Figure 34 Cells cultured with many free edges release ATP independently of cell rupture

Assays for quantities of A. LDH or B. ATP in conditioned media from HCLE cells cultured on plastic strips or

controls.

4.2.2 A drug screen implicates gap junction hemichannels in ATP release.

A number of conduits for damage-free ATP release from cells have been identified including

ABC-family transporters, fusion and release of ATP-loaded vesicles, pannexin channels, and gap

junction hemichannels (Lazarowski et al., 2003; Praetorius and Leipziger, 2009; Schwiebert and

Zsembery, 2003; Spray et al., 2006). To examine mechanisms of transport, I treated cells grown

on plastic strips with various chemical inhibitors and measured ATP release (Figure 35). Drugs

that target ABC-family transporters (glibenclamide and verapamil), nucleotide transporters

(atractyloside), or exocytosis (brefeldin A), did not inhibit ATP release. One inhibitor of gap

junction permeability, 18-glycerrhetinic acid (GA), did not inhibit ATP release, but another,

flufenamic acid (FFA), dramatically reduced ATP release.

54
 Figure 35 A drug screen for inhibitors of

ATP release

ATP concentrations were determined in

conditioned media from cells cultured on

plastic strips following no treatment (NT),

treatment with 10 M glibenclamide (Glib), 10 M verapamil (Ver), 10 M atractyloside (Atra), 10 M brefeldin A

(BFA), 100 M 18-glycerrhetinic acid, or 50 M flufenamic acid (FFA). * indicates significant decrease from NT

condition (p<0.001).

Flufenamic acid (FFA) reversibly inhibits gap junction channels through a mechanism

that does not involve PKC, changes in intracellular calcium concentrations, changes in pH

(Harks et al., 2001), or calcium binding (Eskandari et al., 2002). As shown in Figure 36, FFA

also inhibited ATP release from cells 24 hours after wounding cell sheets by the removal of

agarose barriers. I therefore could not rule out a role for gap junction hemichannels in ATP

release from epithelial edges.

Figure 36 FFA inhibits the persistent release of ATP after

wounding

Assay for quantities of ATP in conditioned media from cells 24

hours after no treatment (NT) or wounding (W). 50 M FFA was

added to cells 30 min prior to measurements.

If HCLE cells secrete ATP through gap junction hemichannels, they are expected to

express functional intercellular gap junctions. This was tested by examining cell coupling with

the gap junction-permeable dye Lucifer yellow (LY) (Abbaci et al., 2008; Stewart, 1978). The

55
dye was introduced into monolayers of HCLE cells by scraping the cells with a pipet tip and

allowing the dye to enter damaged cells and diffuse through gap junctions into neighboring cells

during a five-minute incubation. Cells were simultaneously incubated with rhodamine-

conjugated dextran, which is excluded from gap junctional transport due to its large size.

Fluorescence microscopy clearly showed LY- and rhodamine-dextran-positive cells adjacent to

the scratch, but only LY was found in cells distal from the scratch, indicating the presence of

functional gap junctions. When cells were incubated with 50 M FFA prior to dye loading, LY-

positive cells were restricted to the rhodamine-dextran positive cells, indicating that FFA can

inhibit dye coupling through gap junctions. Incubations with 100 M GA however did not

inhibit dye coupling, indicating that this compound may not inhibit gap junction permeability in

my hands. This may explain the lack of effect of GA relative to FFA in the drug screen (Figure

35).

FFA can inhibit permeability of channels formed by not only connexins, but also by

pannexins, a recently described family of proteins that is structurally homologous to connexins.

Like connexins, pannexins form plasma membrane channels (pannexons) that are permeable to

small molecules such as ATP, but unlike connexons, they are principally non-junctional channels

(D'Hondt et al., 2009; Panchin et al., 2000). Permeability of connexons is more sensitive to FFA

than another inhibitor, carbonexolone (CBX), while pannexons are more sensitive to CBX

(Bruzzone et al., 2005; D'Hondt et al., 2009). As shown in Figure 37, ATP release from cells

cultured on plastic strips was more sensitive to FFA than to CBX. These results are consistent

with the hypothesis that connexin hemichannels mediate ATP release after wounding.

56
 Figure 37 ATP release is sensitive to

inhibitors of connexons and pannexons

ATP concentrations were determined in

conditioned media from cells cultured on plastic

strips following no treatment (NT), treatment

with 5, 50 or 500 M carbonexolone (CBX), or

5, 50, or 500 M flufenamic acid (FFA).

4.2.3 Connexin mimetic peptides do not inhibit ATP release.

I next examined ATP release in the presence of inhibitors that more specifically target

connexons. Connexin mimetic peptides have been designed and are used routinely to inhibit

permeability of gap junctions and connexin hemichannels (Evans and Boitano, 2001; Evans et

al., 2006; Kwak and Jongsma, 1999; Leybaert et al., 2003). These peptides, 10-15 amino acids

in length, bind to extracellular regions specific to particular connexins, and mimic

connexin/connexin binding at gap junctions. Long term (>2 hour) treatment of cells with these

peptides can prevent the formation of gap junctions, but short-term treatment (10-30 minutes)

can prevent hemichannel opening and ATP release, perhaps by functioning as a plug in the

connexon pore (Leybaert et al., 2003).

I investigated the effects of the mimetic peptides Gap26 and Gap27 (Boitano and

Evans, 2000) on the release of ATP from cells cultured on strips. I could not use Gap27 because

it was soluble only at a low pH that interfered with the bioluminescence assay. Treatments with

57
Gap26, however, were feasible. As shown in Figure 38, incubation with 300 M of Gap26 for

30 min, 3 hours, or 24 hours did not inhibit ATP release from cells cultured on strips. However,

LY dye transfer was unaffected by 3-hour treatment with Gap26, indicating that the Gap26 may

not have successfully blocked connexon permeability. Therefore, I cannot conclude from these

mimetic peptide studies whether connexons contribute to ATP release.

Figure 38 The mimetic peptide Gap26 does not inhibit ATP release

Assay for quantities of ATP in conditioned media from cells cultured on plastic strips or controls. As indicated,

cells were incubated with 300 M of Gap26 and media were assayed after A. 30 min, B. 2 hr, and C. 24 hr.

4.3 DISCUSSION

Wounding is known to cause cell damage that results in the rapid release of intracellular

components such as ATP. In this chapter, I have shown that wounding also causes the persistent

release of ATP independently of cell damage. I have also conducted preliminary studies to

identify the mechanism by which ATP is secreted. ATP secretion was inhibited by treatment

with the gap junction inhibitor FFA, but not by treatment with the connexin mimetic peptide

Gap26.

58
 Using two distinct damage-minimizing models, I have made the novel observation that

ATP is persistently released from cells near the wound edge. In migrating neutrophils, ATP has

been found to be released from the leading edge of the cell surface (Chen et al., 2006), but

whether ATP is released from the leading edge of migrating epithelial sheets remains to be

determined. Various imaging techniques based on performing bioluminescence assays in situ

under dark-field microscopy could shed light on this issue (Beigi et al., 1999; Wang et al., 2000).

In both wounding models used, concentrations of ATP in conditioned media increased

four-fold, from roughly 25 nM to 100 nM. Because ATP activates the EGFR in HCLE and other

epithelial cells in the 5-50 M range (concentrations that are typically observed in conditioned

media minutes after wounding), the persistent release of ATP may have functions that are

independent of EGFR signaling. Low nM concentrations of extracellular ATP are sufficient to

induce type-2 purinergic receptor activation and stimulate increases in intracellular calcium

concentrations, but the precise role of persistent ATP release in facilitating wound healing is

unclear (Abbracchio et al., 2006; Geyti et al., 2008).

Extracellular ATP concentrations in the nM range are consistent with a mechanism of

ATP release that is independent of cell damage. My experiments do not support a role for ABC-

family transporters, nucleotide transporters, or exocytosis of ATP-loaded vesicles because their

inhibitors did not block ATP release in my studies as they have in others (Knight et al., 2002;

Lazarowski et al., 2003; Maroto and Hamill, 2001). ATP release was, however, sensitive to

inhibition by FFA and CBX, which block permeability of channels formed by the connexin and

pannexin protein families. Inhibition of ATP release was more sensitive to FFA than to CBX,

suggesting that release is a function of connexon, not pannexon, permeability. Incubations with

the connexin mimetic peptide Gap26 did not inhibit ATP release (Figure 38) or LY dye transfer,

59
so these studies should be repeated in the future using higher purity peptides. Gap26 can bind

and inhibit Cx43, the dominant connexin expressed in the corneal epithelium, but it is unknown

whether it inhibits the actions of other corneal connexins (Shurman et al., 2005; Williams and

Watsky, 2002). Based on amino acid sequence analysis, Gap26 is expected to inhibit all corneal

connexins except Cx30.

Although I cannot conclude that wounding stimulates ATP release through connexin

hemichannels, these preliminary studies can serve as a foundation for more decisive studies. For

example, imaging of connexon-permeable dyes (Stout et al., 2002), approaches involving loss-

of-function (siRNA or dominant-negative) methodologies in HCLE cells, and gain-of-function

methodologies in HeLa cells (that do not express connexins or pannexins) could determine

whether the persistent release of ATP is indeed regulated by connexin hemichannels.

60
5.0 CHARACTERIZATION OF THE ATP/PLD2-INDEPENDENT MECHANISM OF

WOUND-INDUCED EGFR ACTIVATION: ROLES OF SRC-FAMILY KINASES AND

PYK2

5.1 INTRODUCTION

Cells at the edges of wounds undergo profound phenotypic changes and acquire the ability to

migrate rapidly (Block and Klarlund, 2008; Farooqui and Fenteany, 2005; Fenteany et al., 2000;

Omelchenko et al., 2003; Poujade et al., 2007; Tamada et al., 2007). Activation of the EGFR is

absolutely required for these changes to occur (Block et al., 2004; Zieske et al., 2000), and I have

previously identified the presence of dual mechanisms of wound-induced EGFR activation: one

that depends on signaling by extracellular ATP that is dispensable for wound healing, and

another that functions specifically in cells near the wound edge ((Block and Klarlund, 2008) and

Chapters Two and Three).

Recent evidence indicates that Src-Family Kinases (SFKs) control wound-induced EGFR

activation by regulating EGFR ligand shedding after wounding (Block and Klarlund, 2008; Xu et

al., 2006). SFKs contain Src-Homology-2 and -3 (SH2 and SH3) domains that bind to consensus

phosphotyrosine and proline-rich domains, respectively, which can lead to activation by loss of

inhibitory intramolecular interactions (Roskoski, 2005). The signals that control wound-induced

SFK activation, and ultimately stimulate EGFR signaling and healing, are not known.

61
 The focal adhesion kinase (FAK) family of non-receptor tyrosine kinases includes FAK

and Pyk2 (a product of the PTK2B gene; also called Related Adhesion Focal Tyrosine Kinase

(RAFTK) and Cell Adhesion Kinase (CAK)), and is known to bind and activate SFKs

(Avraham et al., 2000; Dikic et al., 1996; Schlaepfer et al., 1999). Also, various stimuli that cells

encounter near the wound edge, such as integrin/matrix interactions, physical stretch, and

calcium influx are known to activate FAK and/or Pyk2 (Assoian and Klein, 2008; Avraham et

al., 2000). In this chapter, I have tested the hypothesis that after epithelial wounding, the FAK

family stimulates SFK activity, which in turn activates the EGFR and promotes motility.

Portions of this chapter have been submitted for publication (see Appendix A).

5.2 RESULTS

5.2.1 Wounding stimulates EGFR ligand release and EGFR activation in the absence of

extracellular ATP.

EGFR activation in cells near the wound edge is regulated by extracellular ATP and by an

extracellular ATP-independent mechanism (Block and Klarlund, 2008). To characterize the

mechanism of EGFR activation in the ATP/PLD2-independent pathway, I have used a model for

detecting signaling in wound-proximal cells (Figure 5) and performed most of the following

experiments in the presence of 25 U/ml apyrase. I have previously shown that similar amounts

of apyrase hydrolyze extracellular ATP to undetectable levels (Figure 9) and inhibit activation of

PKC (Figure 10) and PLD (Figure 19) in cells near the wound edge.

62
 Inclusion of the general protease inhibitor GM 6001 inhibited activation of the EGFR

after wounding (Figure 39), although it did not inhibit activation by exogenous ligand (Block et

al., 2004), which is consistent with the notion that EGFR activation is the result of a proteolytic

event. Also, the LA1 antibody, which blocks activation of the EGFR by extracellular ligands,

and neutralizing anti-AR antibodies inhibited EGFR activation (Figure 39). Together with the

previous observation that AR is released in the presence of apyrase (Figure 10) these data

support that the extracellular ATP-independent pathway activates the EGFR though a triple

membrane-passing mechanism (Figure 4).

Figure 39 EGFR activation by proteolytic ligand

release in the absence of extracellular ATP

Western Blots for activation of EGFR after no

treatment (NT), wounding (W), and/or treatments

with 25 U/ml Apyrase (APY), 1 M GM 6001

(GM), 20 g/ml non-immune serum (NIS), 20

g/ml anti-EGFR antibody (LA1) and 20 g/ml

AR neutralizing antibody (AR).* indicates significant decrease from wounded controls (p<0.001)

5.2.2 Src-family kinases regulate EGFR activation near the wound edge.

The Src-family kinases (SFKs) have been suggested to act upstream of EGFR activation after

wounding (Xu et al., 2006) and after treatment with ATP (Chapter Three). To examine their

possible role in the ATP/PLD2-independent pathway, I first noted that elimination of

extracellular ATP with apyrase did not abolish activation of SFK as monitored by

immunoblotting with an antibody that recognizes the SFKs phosphorylated on Tyr-419

63
(Roskoski, 2005) (Figure 40A). Blocking EGFR kinase activity with tyrphostin AG 1478 did not

block SFK activation in HCLE cells, in agreement with the notion that SFKs are upstream of

EGFR activation (Figure 40B). To assess whether SFKs might also be instrumental in activation

of the EGFR in the ATP/PLD2-independent pathway, I incubated cells with two structurally

dissimilar SFK inhibitors that are known to have distinct non-specific interactions (Bain et al.,

2007). Inclusion of either PP2 or Src Kinase Inhibitor-I (SKI) inhibited activation of the EGFR

and the release of AR in the presence of apyrase (Figure 40C and D), although they did not

inhibit EGFR activation by exogenous ligand (Figure 28B), indicating that SFKs are required for

EGFR activation in this pathway.

Figure 40 SFKs regulate wound-induced EGFR activation in the absence of extracellular ATP

A. Western Blots and densitometry for activation of SFK 5 min after no treatment (NT), wounding (W), and/or

treatment with 25 U/ml Apyrase (APY). * indicates significant increases from unwounded controls (p<0.001). B.

Cells were treated as in A but 1 M AG 1478 (AG) was added 30 min prior to stimulations. C. Activation of EGFR

64
was monitored in the presence of 1 M Src Kinase Inhibitor-I (SKI) or 10 M PP2. D. AR quantities in conditioned

media from cells treated as in C were determined. For C and D, * indicates significant decrease from wounded

controls (p<0.001).

5.2.3 Pyk2, but not FAK, is activated rapidly after wounding, independently of signaling

by extracellular ATP, EGFR, or SFK.

To determine whether the Focal Adhesion Kinase family plays a role in initiating SFK and

EGFR activation, I first tested whether FAK or Pyk2 is activated after wounding. Pyk2, but not

FAK, was activated five minutes after wounding as determined by Western Blotting for an

autophosphorylation site on either kinase (Figure 41A). Pyk2 and FAK activation was observed

after treatment of cells with 100 nM of the phosphatase inhibitor sodium orthovanadate (Figure

41A), indicating the immunoblotting procedure was capable of detecting both kinases when

autophosphorylated. Since FAK was not activated detectably after wounding, I concluded that

FAK signaling is not likely to regulate SFK activity in this context, so I therefore focused on

Pyk2.

I determined that Pyk2 activation proceeds independently of extracellular ATP because

Pyk2 activation was not inhibited by 25 U/ml apyrase (Figure 41B). In parallel studies, 25 U/ml

apyrase did inhibit wound-induced activation of PKC, indicating that the treatment was effective

(Figure 41C). In some systems, Pyk2 is activated downstream of the EGFR (Park et al., 2007;

Schauwienold et al., 2008; Shi and Kehrl, 2004). However, activation of Pyk2 was not reduced

in the presence of 100 nM of the EGFR tyrosine kinase inhibitor AG 1478, which effectively

blocked EGFR autophosphorylation (Figure 41B). Pyk2 contains tyrosine residues that can be

phosphorylated by the SFKs, and some studies indicate that SFK-mediated phosphorylation

65
regulates Pyk2 activity (Dewar et al., 2007; Duong et al., 1998; Sieg et al., 1998). However,

wound-induced Pyk2 autophosphorylation was unaffected by treatment with 1 M of the SFK

inhibitor SKI (Figure 41D). The SFK inhibitor blocked EGFR phosphorylation of Tyr-845,

which is catalyzed by the SFKs, demonstrating the efficacy of the inhibitor (Figure 41D).

Together, these results indicate that Pyk2 is activated independently of the EGFR and SFKs.

Figure 41 Pyk2, not FAK, is activated rapidly after wounding

A. HCLE cells were given no treatment (NT) or were wounded (W) and incubated for 5 minutes and cell extracts

were analyzed by Western Blotting for Pyk2 phosphorylated on Tyr-402 (p-Pyk2) or for FAK phosphorylated on

Tyr-397 (p-FAK). As a positive control for the antibodies, 100 nM Na3VO4 was added to the cells for 30 minutes

prior to analysis (Van). B. In the presence of 25 U/ml apyrase, cells received no additional treatment (NT) or were

wounded (W) and incubated for 5 minutes. Where indicated, cells were treated with 100 nM of AG 1478 (AG) for

30 min prior to and during wounding. Western Blots were probed for p-Pyk2 and were then stripped and re-probed

for total levels of Pyk2 and EGFR phosphorylated on Tyr-1173 (p-EGFR). The ratio of p-Pyk2:Pyk2 was

determined by densitometry and * denotes significant differences from unwounded groups (p<0.001). C. Cells

66
were treated as indicated and Western Blots were probed for phosphorylated MARCKS (p-MARCKS), which is a

marker of PKC activity. D. Cells were treated as in B, except 1 M of Src Kinase Inhibitor-I (SKI) was added for 30

min prior to and during wounding, where indicated. The blots were analyzed with an antibody that recognizes the

EGFR phosphorylated on Tyr-845 (p-EGFR-Y845). Densitometry was performed as in B.

5.2.4 C-Src is activated and associates with Pyk2 after wounding.

Since the pharmacological studies suggested that Pyk2 may be activated upstream of SFK, I

proceeded with various studies to determine whether Pyk2 signaling directly causes SFK

activation after wounding, the first of which was to determine whether Pyk2 and SFKs interact.

Pyk2 autophosphorylation on Tyr-402 presents binding sites for SFK SH2 domains (Dikic et al.,

1996). Thus, SFK can be activated by binding to autophosphorylated Pyk2.

I was able to immunoprecipitate the major SFK isoforms c-Src and Fyn from my cells

(Figure 42) and analyze their activation states with an antibody that recognizes activated forms

of the SFKs. Both c-Src (Figure 43A) and Fyn (Figure 43B) were activated within five minutes

after wounding. SFKs are activated downstream of the EGFR in some systems (Abram and

Courtneidge, 2000), however the presence of 100 nM tyrphostin AG 1478, which effectively

blocked EGFR activity (Figure 41B), had no effect on c-Src or Fyn activation, indicating that c-

Src and Fyn are activated upstream of the EGFR.

67
Figure 42 The SFK isoforms c-Src and Fyn were immunoprecipitated from HCLE cells

Depletion of SFK isoforms from cell extracts is shown by Western Blots for A. c-Src and B. Fyn before (Input) and

after immunoprecipitations (IP).

Figure 43 C-Src and Fyn are activated after wounding

Cells received no treatment (NT) or were wounded (W) and incubated for 5 minutes; all conditions in the presence

of 25 U/ml apyrase. Where indicated, cells were treated with 100 nM of the EGFR inhibitor AG 1478 (AG) for 30

min prior to and during wounding. The extracts were precipitated with A. c-Src and B. Fyn antibodies, and the

precipitates were subjected to Western Blotting for SFK phosphorylated on Tyr-419 (p-SFK) and subsequently for

total levels of c-Src or Fyn. Ratios were determined by densitometry and * denotes significant differences from

unwounded groups (p<0.001).

68
 The SFKs can be activated by binding to Pyk2 after it is autophosphorylated (Dikic et al.,

1996). Analysis of immunoprecipitates of c-Src showed the presence of increased levels of Pyk2

after wounding, suggesting an increased level of association of the two molecules (Figure 44A).

The specificities of the signals were confirmed by performing mock precipitations with non-

immune IgG (Figure 44B). The interaction did not depend on EGFR activation since co-

immunoprecipitation was not blocked by the presence of tyrphostin AG 1478 (Figure 44A). A

similar analysis with c-Fyn was not possible because the signals were too faint. Together, these

results indicate that wounding stimulates activation of multiple SFK isoforms and promotes

formation of a c-Src/Pyk2 signaling complex.

Figure 44 C-Src interacts with Pyk2 after wounding

A. Cells received no treatment (NT) or were wounded (W) and incubated for 5 minutes; all conditions in the

presence of 25 U/ml apyrase. Where indicated, cells were treated with 100 nM of the EGFR inhibitor AG 1478

(AG) for 30 min prior to and during wounding. The extracts were precipitated with c-Src antibodies and the

precipitates were subjected to Western Blotting for Pyk2 phosphorylated on Tyr-402 (p-Pyk2) and subsequently for

total levels of c-Src. Ratios were determined by densitometry and * denotes significant differences from unwounded

69
groups (p<0.01). B. Lysates were subjected to immunoprecipitation with anti-c-Src antibodies or non-immune

mouse IgG; immunoprecipitates were subjected to Western Blotting for p-Pyk2. Over-exposed film shows the

absence of signal from the non-immune control.

5.2.5 Pyk2 signaling is necessary for SFK and EGFR activation after wounding.

To determine whether Pyk2 is necessary for wound-induced SFK activation, I employed multiple

approaches to modulate Pyk2 signaling. I initially examined the effects of reducing Pyk2

expression with siRNA prior to wounding. Cells were transfected with 10 nM of a pool of Pyk2-

targeted siRNA oligonucleotides before wounding. Pyk2 siRNA transfection caused 70%

reduction in Pyk2 expression compared to controls and resulted in inhibition of wound-induced

SFK and EGFR activation (Figure 45). As a confirmation and to better control for off-target

effects, I repeated these studies using 10 nM of a single Pyk2-targeted siRNA oligonucleotide.

The Pyk2 siRNA caused approximately 75% reduction in Pyk2 expression (Figure 46A) and

inhibited significantly SFK (Figure 46B) and EGFR (Figure 46C) activation after wounding.

Treatment of Pyk2 siRNA-transfected cells with EGF yielded equivalent EGFR activation as

treatment of control-transfected cells (Figure 46D), indicating that the siRNA did not interfere

with the normal kinase activity of the EGFR.

Figure 45 Knockdown of Pyk2 with siRNA inhibits SFK

and EGFR activation after wounding

HCLE cells mock-transfected or transfected with a pool of

Pyk2 siRNA oligonucleotides were wounded as indicated.

25 U/ml apyrase was present in all conditions. Western

Blots were probed for activation of SFK and EGFR and for

verification of Pyk2 knockdown.

70
Figure 46 Knockdown of Pyk2 with siRNA inhibits SFK and EGFR activation after wounding but not after

stimulation with EGF

HCLE cells transfected with 10 nM of control or Pyk2 siRNA were treated with apyrase and were wounded as

indicated. Western Blots were probed for A. total levels of Pyk2 and Actin, B. SFK phosphorylated (p-SFK) and

non-phosphorylated (SFK) on Tyr-419, and C. EGFR phosphorylated on Tyr-1173 (p-EGFR) and total levels of

EGFR. Ratios were determined by densitometry and * denotes significant decrease from wounded controls

(p<0.001). D. HCLE cells transfected with control or Pyk2 siRNA were treated with 10 ng/ml EGF for 5 min where

indicated prior to Western Blotting and densitometry.

71
 As a complement to these studies, I tested whether Pyk2 is necessary for wound-induced

SFK and EGFR activation by wounding sheets of HCLE cells that express Pyk2-Related Non-

kinase (PRNK), which is a truncated version of Pyk2 that consists of the C-terminal 120 amino

acids and lacks the kinase domain. It can act in a dominant-negative fashion to inhibit

endogenous Pyk2 activation and signaling (Xiong et al., 1998). PRNK (~30 kDa) was clearly

expressed in HCLE cells infected with an adenovirus coding for PRNK, and its presence blocked

Pyk2 activation after wounding as expected (Figure 47A). Importantly, activation of SFKs

(Figure 47B) and the EGFR (Figure 47C) were also significantly inhibited by the presence of

PRNK. EGF-induced EGFR activation was not inhibited by PRNK expression (Figure 47D),

indicating that PRNK did not affect the EGFR directly.

I also examined the effects of blocking Pyk2 signaling on the individual SFK isoforms.

C-Src activation was inhibited when Pyk2 signaling was disrupted by siRNA (Figure 48A) or

PRNK (Figure 48B). Similarly, activation of the Fyn isoform was inhibited by Pyk2 siRNA

(Figure 48C) or PRNK (Figure 48D). Therefore, although I could not determine the degree to

which Fyn and Pyk2 interact after wounding, Pyk2 signaling does appear to be necessary for Fyn

activation. Together, the results from the studies with siRNA and PRNK support that Pyk2

mediates activation of the SFKs and the EGFR after wounding.

72
Figure 47 Expression of a Pyk2 dominant-negative mutant inhibits SFK and EGFR activation after wounding

but not after stimulation with EGF

HCLE cells infected with control adenovirus or adenovirus coding for PRNK were incubated with apyrase and were

wounded as indicated. Western Blots were probed for A. Pyk2 phosphorylated on Tyr-402 (p-Pyk2), total levels of

Pyk2, PRNK, and Actin, B. SFK phosphorylated (p-SFK) and non-phosphorylated (SFK) on Tyr-419, and C.

EGFR phosphorylated on Tyr-1173 (p-EGFR) and total levels of EGFR. Ratios were determined by densitometry

and * denotes significant decrease from wounded controls (p<0.001). D. HCLE cells infected with control

adenovirus or adenovirus coding for PRNK were treated with 10 ng/ml EGF for 5 min where indicated prior to

Western Blotting and densitometry.

73
Figure 48 Disruption of Pyk2 signaling by siRNA or PRNK inhibits wound-induced activation of c-Src and

Fyn

HCLE cells transfected with 10 nM of control or Pyk2 siRNA (A and C) or infected with control adenovirus or

adenovirus coding for PRNK (B and D) were treated with apyrase and were wounded as indicated. C-Src

immunoprecipitates were immunoblotted for SFK phosphorylated on Tyr-419 (p-SFK) and for total levels of c-Src

(A and B). Fyn immunoprecipitates were immunoblotted for SFK phosphorylated on Tyr-419 (p-SFK) and for total

levels of Fyn (C and D). Ratios were determined by densitometry and * denotes significant decrease from wounded

controls (p<0.001).

74
 To test whether increasing Pyk2 signaling promotes SFK and EGFR activation, I infected

cells with adenovirus encoding wild-type Pyk2, which caused highly increased levels of

activated Pyk2, perhaps because the over-expression overwhelmed Pyk2 phosphatases or other

negative regulators. Pyk2 over-expression caused highly increased levels of activated SFKs

(Figure 49A), which was expected due to the direct association of the two kinases. Importantly,

Pyk2 overexpression also induced significant increases in the levels of activated EGFR (Figure

49B) in agreement with a role for Pyk2 as an upstream activator. The activation of SFKs by

Pyk2 is direct, but many signaling intermediates are required for EGFR activation by SFKs, so

relative to SFKs, the EGFR was only modestly activated by Pyk2 expression.

Figure 49 Over-expression of wild-

type Pyk2 stimulates SFK and EGFR

activation

HCLE cells infected with control

adenovirus (Con.) or adenovirus coding

for Pyk2 were subjected to Western

Blotting for A. Pyk2 phosphorylated on

Tyr-402 (p-Pyk2), total levels of Pyk2, SFK phosphorylated (p-SFK) and non-phosphorylated (SFK) on Tyr-419,

and -actin, and B. EGFR phosphorylated on Tyr-1173 (p-EGFR) and total levels of EGFR. Ratios were

determined by densitometry and * denotes significant increase from controls (p<0.01).

5.2.6 Pyk2 activation is necessary for wound healing.

Since Pyk2 regulates SFK and EGFR activation, I expected that modulation of Pyk2 signaling

would alter healing rates in our cell culture wound healing assay. Transfection with 10 nM of a

75
single Pyk2 siRNA oligonucleotide prior to wounding reduced Pyk2 expression by ~60% and

retarded wound healing by 40% compared to transfection with control siRNA (Figure 50A). The

effect of Pyk2 siRNA on healing was not due to differences in cell division or death because

similar cell densities were observed in control and Pyk2 siRNA-transfected cultures (Figure

50B). Treatment with 10 ng/ml EGF during healing resulted in similar accelerated healing rates

in both control and Pyk2 siRNA-transfected conditions (Figure 50A), indicating that the effects

of Pyk2 siRNA are upstream of the EGFR.

Figure 50 Pyk2 activation is

necessary for wound healing

HCLE cells transfected with 10

nM of control or Pyk2 siRNA

were subjected to wound healing

assays in the presence of 25 U/ml

apyrase. Where indicated, cells

were incubated with 10 ng/ml EGF during healing. A. The percentages of the areas covered during 18 hour periods

of healing are shown. * indicates significant difference from untreated controls (p<0.01). B. Cells were fixed,

stained with DAPI, and the average number of nuclei/mm2 in ten random fields was determined.

Infection of cells with PRNK-expressing adenovirus reduced wound healing by 60%

compared to infection with control virus (Figure 51A). Cell densities were similar in control and

PRNK-infected cultures (Figure 51B). The PRNK mutant used was tagged with the FLAG

epitope, so PRNK expression could be monitored by immunofluorescence microscopy.

Interestingly, in fields where PRNK expression was not induced in all of the cells, the FLAG-

negative cells had migrated further than the FLAG-positive cells (Figure 51C). These data are

76
consistent with the hypothesis that Pyk2 activation is necessary for cell motility. EGF stimulated

healing in PRNK-expressing cultures 20% less than it stimulated healing in controls (Figure

51A), which indicates that PRNK expression in itself causes a slight defect in wound healing.

Pyk2 and PRNK interact with paxillin (Li and Earp, 1997), a component of focal adhesions, so I

monitored localization of paxillin by immunofluorescence confocal microscopy. In both FLAG-

negative and FLAG-positive cells, paxillin was observed in small puncta near the base of the cell

that co-localized with termini of actin fibers (Figure 51D), indicating that PRNK expression does

not cause any major perturbation of focal adhesions.

Figure 51 Pyk2

activation is

necessary for wound

healing

A. Healing and B.

density of cells

infected with control

adenovirus or

adenovirus coding for

FLAG-tagged PRNK

was measured as for

Figure 47. *

(p<0.001) indicates

significant differences

from untreated controls and ** (p<0.05) indicates significant difference from EGF-treated controls.

Immunofluorescence images of cells from A taken with a C. 20X or D. 60X objective. Only PRNK-expressing

cells stain for FLAG.

77
 Since inhibition of Pyk2 signaling by either siRNA or the dominant negative PRNK

resulted in inhibition of healing, I conclude that induction of cell motility in epithelial cells is

dependent on Pyk2 and its downstream signaling to the SFKs and the EGFR.

5.3 DISCUSSION

Activation of the EGFR in cells near the wound edge is a decisive signal for the induction of

motility, but the signals that trigger EGFR activation after wounding are not completely known.

In this chapter, I have shown that EGFR activation is dependent on SFKs that signal to the EGFR

through a triple membrane-passing mechanism (Figure 4). I have also identified Pyk2 as a key

regulator of wound-induced SFK and EGFR activation that promotes cell migration. Evidence

for this conclusion includes: 1, Pyk2 was activated rapidly after wounding, and pharmacological

inhibition of signaling by the SFKs or the EGFR did not abrogate Pyk2 activation; 2, association

of c-Src and Pyk2 was increased after wounding; 3, over-expression of wild-type Pyk2

stimulated EGFR activation, and 4, inhibition of Pyk2 signaling with siRNA or a dominant-

negative mutant inhibited SFK and EGFR activation and healing of wounds in cell sheets.

In the present study, I have observed partial, but not complete inhibition of EGFR

activation and wound healing by Pyk2 siRNA or by expression of PRNK. This may be due to

incomplete inhibition of Pyk2 signaling or alternatively, it may suggest the presence of

additional mechanisms of SFK and EGFR activation after wounding. Wounding a sheet of

epithelial cells is a complex event that challenges cells with numerous stimuli that have been

hypothesized to promote EGFR activation, including exposure to extracellular matrix, physical

stretch, loss of polarity, and loss of physical constraints (Iwasaki et al., 2000; Moro et al., 1998;

78
Vermeer et al., 2003), and full activation of the EGFR may therefore depend on several signals.

Alternate mechanisms of wound-induced EGFR receptor activation are discussed further in

Chapter Six.

The mechanism of wound-induced Pyk2 activation is presently unknown. Pyk2 activity

is sensitive to numerous environmental cues including integrin/matrix binding, calcium

concentration, growth factors and cytokines, reactive oxygen species, and actin cytoskeletal

dynamics (Avraham et al., 2000; Schlaepfer et al., 1999). Within minutes of wounding, calcium

influx, reactive oxygen species generation, and integrin adhesions have all been observed in cells

near the wound edge (Klepeis et al., 2001; Nikolic et al., 2006; Sammak et al., 1997; Zaidel-Bar

et al., 2003), but the extent to which these signals contribute to Pyk2, SFK, or EGFR activation

in the present system is unclear.

Some evidence indicates that Pyk2 activation is stimulated by EGFR signaling (Park et

al., 2007; Schauwienold et al., 2008; Shi and Kehrl, 2004). I have observed Pyk2 activation in

HCLE cells treated with EGF, which suggests the presence of positive feedback loops in the

wounded epithelium. Indeed, inhibition of EGFR signaling with tyrphostin AG 1478 resulted in

the slight reduction of Pyk2 activation after wounding (Figure 41), although these results were

not statistically significant.

Many studies investigating GPCR signaling have suggested through indirect means that

Pyk2 mediates EGFR transactivation (Burdick et al., 2006; Kanno et al., 2003; Montiel et al.,

2007; Park et al., 2006), although some have indicated that Pyk2 and EGFR activation are

unrelated events (Kodama et al., 2002; Zwick et al., 1999). In this report, I have tested the

hypothesis directly that Pyk2 signaling is necessary for wound-induced EGFR activation by

using Pyk2-targeted siRNA oligonucleotides (Figure 46) and by expression of the Pyk2

79
dominant negative, PRNK (Figure 47). Two other studies have also tested directly whether Pyk2

signaling is necessary for EGFR phosphorylation: EGFR transactivation by GPCR agonists was

inhibited in Pyk2-deficient fibroblasts (Andreev et al., 2001), and EGFR transactivation by

thiazolidinediones was inhibited by Pyk2-targeted siRNA in a liver epithelial cell line (Dewar et

al., 2007). Therefore, data presented here add to the mounting evidence for an important

connection between Pyk2 and the EGFR. This generates interest in Pyk2 as a therapeutic target

for not only wound healing, but also EGFR-related pathologies such as certain cancers and

cardiac hypertrophy (Behmoaram et al., 2008; Hirotani et al., 2004; Iiizumi et al., 2008; Menashi

and Loftus, 2009; Roelle et al., 2008; Sun et al., 2008).

Most studies are in accord with a positive role for Pyk2 in promoting motility, as I have

shown here in HCLE cells. Decreasing Pyk2 expression inhibited motility of vascular smooth

muscle cells (Soe et al., 2009), glioma cells (Lipinski et al., 2005) and macrophages (Owen et al.,

2007). Furthermore, Pyk2 over-expression increased motility of glioma (Lipinski et al., 2005)

and hepatocellular carcinoma (Sun et al., 2008) cells and expression of a constitutively active

Pyk2 mutant enhanced motility of endothelial (Kuwabara et al., 2004) and breast cancer (Zrihan-

Licht et al., 2000) cells. However, a few studies are at variance with this conclusion: Pyk2

siRNA stimulated motility of prostate epithelial cells (de Amicis et al., 2006), and a kinase-dead

Pyk2 mutant had no effect on motility of breast cancer cells (Zrihan-Licht et al., 2000). Pyk2

knockout mice (Okigaki et al., 2003) would be an excellent model to further examine the role of

Pyk2 in epithelial migration and repair. My conclusion that Pyk2 is an upstream activator of the

EGFR provides a rationale for a pro-migratory role in cell motility because the EGFR is widely

recognized to stimulate cell migration.

80
5.3.1 Summary and model

In summary, the results presented here identify Pyk2 activation as a critical signal that leads to

SFK and EGFR activation in cells near the wound edge. By stimulating SFKs and the EGFR,

Pyk2 activation is necessary for healing of epithelial wounds (see diagram in Figure 52).

Figure 52 A hypothetical model for the role of Pyk2 in wound-induced EGFR activation and cell migration

See text for details. Pyk2, the focal adhesion kinase Pyk2; SFK, Src-family kinase; MMP, matrix metalloprotease;

ADAM, a disintegrin and metalloprotease; EGFR, epidermal growth factor receptor.

81
 6.0 EPITHELIAL FREE EDGES TRIGGER ACTIVATION OF THE EGFR

6.1 INTRODUCTION

Wounding is a complex event that presents epithelial cells with numerous stimuli that could

promote EGFR activation and the ensuing transition to the motile state. The relative importance

of these stimuli may vary based on the type of wound, the distance from the wound, or the time

after wounding. For example, as I have shown in Chapters Two and Four, ATP is released

rapidly after wounding and can stimulate EGFR activation in cells far from the wound, but hours

later, ATP is secreted at levels too low to induce EGFR activation. As another example, alkali

burns disrupt the corneal basement membrane differently than scratch wounds, thereby altering

cell/matrix interactions that may affect signaling (DeMali and Burridge, 2003; Guo and

Giancotti, 2004; Moro et al., 1998; Saika et al., 1993). Whatever the manner of infliction or time

after incidence, a common factor shared by all wounds is the presence of a physically

unconstrained, or free, edge.

Free edges in epithelial sheets are not exclusive to wounds but are in fact found during

embryonic development. In a mammalian model of epithelial morphogenesis, eyelid closure in

mice, epithelial sheet movement is dependent on the proteolytic release of HB-EGF that activates

the EGFR (Mine et al., 2005). Therefore, not only are the cellular processes that regulate

epithelial movements during morphogenesis and wound healing similar, but the signals that

82
induce this motility are similar as well (Friedl and Gilmour, 2009; Jacinto et al., 2001; Martin

and Parkhurst, 2004).

Identifying signaling mechanisms that regulate the activation of the EGFR has been a

major focus of recent wounding research (Yu et al., 2009). The majority of this research has

focused on signaling that occurs minutes after wounding, but studies of the epidermal and

corneal epithelia have indicated that EGFR activation persists as long as a free edge is present

(Shirakata et al., 2005; Stoscheck et al., 1992; Zieske et al., 2000). To study signaling that

results from free edges, I have employed a model for comparing cell cultures containing

numerous edges with cultures containing no edges (see Figure 7 and Methods 8.6). Under the

culture conditions, cell damage and access to unoccupied adjacent extracellular matrix is

negligible. Using this and other models, I have tested the hypothesis that a free edge alone is

sufficient to trigger EGFR activation. Portions of this chapter have been submitted for

publication (see Appendix A).

6.2 RESULTS

6.2.1 Chronic EGFR activation is necessary for epithelial migration.

Studies in the epidermal and corneal epithelia have indicated that wounding stimulates prolonged

activation of the EGFR, which may lead to its downregulation near the wound edge (Shirakata et

al., 2005; Stoscheck et al., 1992; Zieske et al., 2000). To examine the kinetics of EGFR

activation near the edges of wounded HCLE cell sheets, I monitored EGFR activation for hours

after the removal of agarose droplets (Figure 53A). Rapid activation of the EGFR was followed

83
by downregulation of total levels of the receptor. 24 hours after wounding, the ratio of active to

total EGFR remained elevated relative to controls (Figure 53B). Similarly to the EGFR,

activation of ERK1/2 persisted for hours (Figure 53A and B).

Figure 53 Wounding causes persistent EGFR and ERK1/2 activation

A. HCLE cell sheets were wounded by the removal of agarose droplets and the activation of EGFR and of ERK1/2

was monitored for 24 hr by Western Blotting. B. Ratios were determined by densitometry of Western Blots from 5

minutes (5) or 24 hours (24h) after wounding. * indicates significant difference from unwounded controls

(p<0.001).

The EGFR inhibitor AG 1478 blocked ERK1/2 activation at both the early and late time

points (Figure 54; this figure, as well as figures 55, 56B, and 64 were not prepared by the author,

see acknowledgments), indicating that the EGFR is actively signaling even when total levels are

downregulated. In wound healing assays, addition of AG 1478 at any time during healing

slowed cell migration (Figure 55). These data indicate that chronic EGFR activation is necessary

for the migration of epithelial sheets.

Figure 54 EGFR signaling regulates ERK1/2

activation after wounding

Western Blotting for activation of ERK1/2 at 0 or 5

minutes or 24 hours after wounding. As indicated, 1 M of the EGFR inhibitor AG 1478 (AG) was added 30 min

prior to cell lysis. Figure prepared by Mike Tolino.

84
 Figure 55 EGFR activity is necessary throughout healing

Wounded HCLE cell sheets were fixed at 0, 5, 10, 15, or 30

hours. For comparison, 1 M of the EGFR inhibitor AG

1478 (AG) was added at the indicated times, and healing was

allowed to progress over the full 30 hours before fixation.

Figure prepared by Mike Tolino.

6.2.2 A cell culture model for detecting signaling by free edges.

Many of the stimuli associated with wounding could cause chronic EGFR activation in cells near

the wound edge. We developed a model to examine the consequences of a stimulus that is by

definition ubiquitous in wounds, the presence of a free edge (see Figure 7, and Methods). To

produce a cell culture with many free edges, HCLE cells were cultured on thin plastic strips

above a non-adhesive surface of polyHEMA. In xz sections generated by confocal microscopy, I

observed that cells at the free edge were not physically constrained by the polyHEMA and that

their membranes extended over the edges (Figure 56A).

Rates of thymidine uptake were reduced similarly in cells grown on strips and in control

cultures when they reached confluence (Figure 56B). Levels of lactate dehydrogenase in the

supernatants of cells grown in the two conditions were low and similar, indicating that growth of

cells on the strips does not promote cell lysis (Loo and Rillema, 1998) (Figure 34A). These

results indicate that differences in cells grown on strips or controls are unlikely to be due to

differences in proliferation states or cell lysis.

85
Figure 56 Cells grown on plastic strips have a physically unconstrained edge and slow proliferation when they

reach confluence

A. Cross-sections of cells grown on plastic strips were generated by confocal microscopy. The plastic strips and

polyHEMA were labeled with fluorophores (shown in green and blue, respectively), and the cells were labeled with

the membrane dye Vybrant DiD (depicted in red). B. 3H-thymidine uptake by cells grown on plastic strips or

controls at sparse or confluent densities. No significant differences were detected.

6.2.3 EGFR activation requires physically unconstrained edges.

I performed Western Blotting on extracts from HCLE cells cultured on plastic strips and found

that relative to controls, the EGFR was down-regulated but there was an increased ratio of active

to total EGFR (Figure 57A and B). These results are similar to those observed in cells near the

edge 24 hours after wounding (Figure 53). I also analyzed Madin-Darby canine kidney (MDCK)

cells grown on plastic strips and found increased EGFR phosphorylation, although EGFR down-

regulation was less pronounced (Figure 58). ERK1/2 were also activated in cells cultured on

plastic strips (Figure 59A) and this activation was dependent on EGFR signaling because it was
86
abrogated by the inclusion of the EGFR inhibitor AG 1478 (Figure 59B). Similar results were

obtained when agarose was used as the non-adhesive substrate instead of polyHEMA.

Figure 57 Activation of the EGFR by free edges in HCLE cell

sheets

A. Immunoblot of extracts of HCLE cells with an antibody against

EGFR phosphorylated on Tyr-1173. The blots were stripped, and re-

probed for total amounts of the EGFR and -actin as a load control.

B. Quantitation of Western Blots by densitometry. * indicates

significant differences from controls (p<0.001).

Figure 58 Activation of the EGFR by free edges in MDCK cell

sheets

A. Immunoblot of extracts of MDCK cells was performed as for

Figure 53. B. Quantitation of Western Blots by densitometry. *

indicates significant differences from controls (p<0.001).

Figure 59 Activation of ERK1/2 by free

edges

A. Western Blotting and densitometry for

active ERK1/2 and -actin as a load control.

* indicates significant difference from

controls (p<0.001). B. Cells cultured on

strips or controls were treated for 30 min

with 1 M of the EGFR inhibitor AG 1478

(AG). * indicates significant difference from untreated cells on strips (p<0.001).

87
 To determine how far from an edge activation occurs, cells were seeded on polyHEMA-

coated plates that were half-covered with plastic, and therefore contained a single edge (see

Methods 8.7). Immunoblotting extracts of cells located at various distances from the edge

revealed that activation occurred locally within the most proximal 5 mm from the edge (Figure

60). Treatment of cells cultured on re-dissolved plastic with the MAPK inhibitor UO126 led to

quenching of pERK1/2 activation as detected by Western Blotting but not as detected by

immunofluorescence microscopy. Therefore, although microscopy could be a useful tool for

examining the penetrance of activation, I concluded that culture on re-dissolved plastic creates

some artifactual signal that subverts the utility of immunofluorescence microscopy.

Figure 60 ERK1/2 activation is limited to near the free edge

Cells were cultured with one free edge by growing on polyHEMA-coated plates

that were half covered in plastic. On ice, 0.5 cm strips were sequentially scraped

away from the edge and were analyzed individually. Densitometry of Western

Blots is shown. * indicates significant differences from all other groups

(p<0.001).

To determine whether a physical barrier can inhibit EGFR activation in strips of

epithelial cells, I analyzed cells grown in thin strips around agarose droplets (see Figure 5 and

Methods). In xz sections generated by confocal microscopy, I observed that cells were

physically constrained by the agarose droplets (Figure 61A). Levels of EGFR activation,

downregulation, and ERK1/2 activation were similar in these and control confluent cultures

lacking agarose (Figure 61B and C). Addition of EGF to cells grown either as uninterrupted

sheets or with agarose droplets resulted in similar levels of EGFR activation (Figure 61D). I

88
conclude that free edges in epithelial cell sheets induce EGFR activation while edges that are

physically constrained by an inert material such as agarose do not.

Figure 61 Edges are physically constrained do not activate the EGFR

A. Cells were labeled with the membrane dye Vybrant DiD (red) and agarose droplets were labeled with

fluorescein (green). The plastic below the cells and agarose droplets was left unlabeled for clarity. B. Western Blots

of extracts of HCLE cells cultured without and with agarose droplets. C. Quantitation of Western Blots by

densitometry. No significant differences from controls were found. D. EGFR activation was monitored after

stimulation of HCLE cells grown either as uninterrupted sheets or around agarose droplets with 10 ng/ml EGF for

10 min.

89
6.2.4 Free edges stimulate EGFR activation through SFK-mediated proteolytic ligand

release.

Wounding activates the EGFR rapidly by the proteolytic release of membrane-bound ligands

(Block et al., 2004; Xu et al., 2004). To determine whether EGFR activation induced by free

edges is regulated by a similar mechanism, I incubated cells with various inhibitors of signaling

by EGFR ligands and monitored activation of ERK1/2, which are downstream of the EGFR.

ERK1/2 activation was abrogated by treatment with GM6001, which inhibits EGFR pro-ligand

shedding, and by the anti-EGFR antibody LA-1, which inhibits extracellular ligands from

binding the EGFR (Figure 62). Therefore, I conclude that free edges activate the EGFR through

proteolytic ligand release.

Figure 62 Free edges activate ERK1/2 through

proteolytic release of EGFR ligands

Western Blots and densitometry for activation of

ERK1/2 in cells cultured on strips after no treatment

(NT), or treatment with 1 M GM negative control

(GM-neg), 1 M GM 6001 (GM), 20 g/ml non-

immune IgG (IgG), or 20 g/ml anti-EGFR antibody

(LA1). * indicates significant difference from untreated

cells on strips.

Src-family kinases (SFKs) regulate EGFR ligand release after wounding ((Block and

Klarlund, 2008) and Chapter Five), so I expected that SFKs also regulate EGFR activation

stimulated by free edges. I observed strong downregulation of SFKs using antibodies that

90
recognize active or inactive forms of the kinases (Figure 63A). Highly activated forms of SFKs

are targeted for degradation, so SFK down-regulation is likely a result of chronic SFK activation

(Ingley, 2008). I next monitored ERK1/2 activation in the presence of two structurally dissimilar

SFK inhibitors that are known to have distinct non-specific interactions, PP2 and Src Kinase

Inhibitor-I (SKI) (Bain et al., 2007). Inclusion of either SFK inhibitor abrogated ERK1/2

activation (Figure 63B). Together, these results suggest that free edges activate the EGFR

through SFK signaling.

Figure 63 Activation of SFKs by free

edges is necessary for ERK1/2 activation

A. Western Blotting and densitometry of

extracts with an antibodies against SFKs and

-actin as a load control. * indicates

significant differences from controls (p<0.001). B. ERK1/2 activation was monitored in cells on strips after 30 min

treatment with 1 M Src Kinase Inhibitor-I (SKI) or 10 M PP2.

6.2.5 Free edges stimulate EGFR/ERK activation independently of signaling by ATP,

Pyk2, or loss of polarity

SFK and EGFR activation after wounding can be regulated by extracellular ATP ((Block and

Klarlund, 2008; Yin et al., 2007), Chapters Two and Three). However, the amount of ATP in

supernatants from cells grown on plastic strips, approximately 100 nM (Figure 34), is too low to

induce measurable activation of the EGFR and ERK1/2 (Yin et al., 2007), and addition of

apyrase did not block activation of the EGFR/ERK1/2 pathway in cells grown on strips (Figure

91
64). These data strongly suggest that activation occurs through a mechanism that is independent

of signaling by extracellular ATP.

Figure 64 Free edges activate ERK1/2 independently of

extracellular ATP signaling

Western Blots and densitometry for activation of ERK1/2 in cells

cultured on strips or controls and treated with 30 U/ml apyrase (Apy) as

indicated.

SFK and EGFR activation after wounding can also be regulated by Pyk2 (Chapter Five).

I did not observe Pyk2 activation in cells grown on strips, relative to controls (Figure 65A).

Furthermore, in preliminary studies, activation of the EGFR and ERK1/2 were unaffected by

knockdown of Pyk2 by siRNA (Figure 65B). These data suggest that free edges stimulate EGFR

activation by a mechanism independent of signaling by Pyk2.

Figure 65 Free edges activate the EGFR

independently of signaling by Pyk2

A. Western Blots for activation of Pyk2 in

cells cultured on strips and controls.

Similar results were obtained in four

independent experiments. B. Activation

of EGFR and ERK1/2 and expression of

Pyk2 in cells cultured on strips and controls was monitored after transfection with 10 nM of Pyk2 siRNA. Due to

splice variants and post-translational modifications, multiple Pyk2 bands can be resolved by Western Blotting.

92
 In the airway epithelium, ErbB2-3, which are related to the EGFR, are activated by a

mechanism that is based on segregation of the apical ligand (heregulin) from the basolateral

receptors (Vermeer et al., 2003). Wounding breaks down the barrier function of the epithelium

allowing heregulin and the receptors to interact. For such a mechanism to operate, the epithelial

cells must provide effective separation of apical and basolateral compartments, which is mainly

due to tight junctions. Specialized procedures are required for epithelial cells to form fully-

developed tight junctions in tissue culture and as expected they were incomplete in HCLE cells

grown on redissolved plastic (Figure 66A). Furthermore, EGF added apically to cells grown

under control conditions potently activated the EGFR in basolateral membranes (Figure 66B).

Figure 66 EGFR ligands are not segregated from receptors

Xy and xz projections were generated by confocal microscopy of HCLE cells cultured on plastic-covered

polyHEMA and stained for A. ZO-1 or B. phosphorylated EGFR. As indicated, cells were treated with 100 ng/ml

EGF for 2 minutes.

93
Since the HCLE cells do not provide an effective barrier between apical and basolateral

compartments, activation of the EGFR by the presence of free edges cannot be due to its

disruption.

6.3 DISCUSSION

In this chapter, I have investigated the signaling caused by an element common to migrating

epithelial sheets found during development in the embryo and wound repair in the adult: the

presence of a free edge. I have shown that the EGFR is persistently activated in cells near the

free edge and that an epithelial sheet requires persistent EGFR activation to migrate. To

determine whether a free edge is sufficient to induce this chronic activation, we developed

methods to compare cultures of cells with many free edges, many constrained edges, and

virtually no edges. I have concluded that free edges trigger EGFR activation based on two key

observations: 1, the EGFR and ERK1/2 were activated similarly in cells near the free edge after

either wounding or culture on plastic strips and 2, constraining the free edge with agarose

blocked activation of the EGFR and ERK1/2.

There are marked similarities in the mechanisms of epithelial movement during the

healing of wounds and during development (Friedl and Gilmour, 2009; Jacinto et al., 2001;

Martin and Parkhurst, 2004). For example, cells near the leading edge are propelled forward by

a combination of lamellipodial crawling and purse-string-like contractions of the actin

cytoskeleton (Danjo and Gipson, 1998; Mitchison and Cramer, 1996; Tamada et al., 2007). For

decades, researchers have speculated that a free edge may be a signal that induces these and

other changes, but there has been little hard evidence (Martin and Parkhurst, 2004; Rand, 1915).

94
Here I have made the observation that free edges induced EGFR activation in not only HCLE

cells, but also in MDCK cells (Figure 59), a prototypical epithelial model, which suggests that

the ability to detect free edges is a universal mechanism that promotes epithelial motility.

I was unable to examine directly whether a free edge induces EGFR signaling acutely

after wounding. In Chapter Five, I observed that blocking signaling by both Pyk2 and

extracellular ATP resulted in only partial inhibition of wound-induced EGFR activation.

Because free edges are created suddenly by the removal of agarose barriers, I cannot exclude the

possibility that the presence of a free edge is a stimulus for rapid EGFR activation as well. If, in

the future, I find molecular or pharmacological approaches that inhibit SFK activation by free

edges (see Chapter Seven), I will be able to use these same approaches to determine whether free

edges contribute to acute EGFR activation after wounding.

I have found that the mechanism of EGFR activation by free edges involves the SFK-

mediated release of EGFR pro-ligands (Figure 63). Other proposed mechanisms of wound-

induced EGFR activation suggest that activation occurs not as a result of increased SFK-

mediated ligand shedding, but rather as a result of basal ligand shedding having a greater effect

due to increased access to receptors (Chu et al., 2005; Tschumperlin et al., 2004; Vermeer et al.,

2003). The observation that a free edge stimulates SFK activation suggests that free edges

actively stimulate signaling upstream of the EGFR.

The observation that edges do not cause activation when they are constrained by agarose

is significant, because it strongly suggests that suppression of activation is not due to any specific

interaction between cells. For instance, contact inhibition of cell locomotion is usually thought to

result from activation of receptors on cells recognizing ligands on adjacent cells (Abercrombie,

1979; Carmona-Fontaine et al., 2008). More likely, activation is repressed in epithelial cell

95
sheets by mechanical forces that the cells exert on one another (Bindschadler and McGrath,

2007; Farooqui and Fenteany, 2005). At unconstrained edges, cells do not encounter opposing

forces, and tensions in the cells would be relaxed. As is discussed further in Chapter Seven,

ongoing research is exploring the possibility that the free edge sensor may involve the actin

cytoskeleton, which is a keen sensor of tensile stress that signals through SFKs (Ingber, 2006;

Ingley, 2008).

6.3.1 Summary and model

In summary, the results presented here indicate that free edges in epithelial cell sheets stimulate

signals that activate SFKs and, subsequently, the EGFR (see diagram in Figure 67).

Figure 67 A hypothetical model for the role of free edges in wound-induced EGFR activation

See text for details. SFK, Src-family kinase; MMP, matrix metalloprotease; ADAM, a disintegrin and

metalloprotease; EGFR, epidermal growth factor receptor.

96
 7.0 CONCLUSIONS

7.1 A MODEL FOR MECHANISMS OF WOUND-INDUCED EGFR ACTIVATION

The goal of the research presented in this dissertation was to gain a better understanding of the

cellular signals that induce migration after epithelial wounding. Specifically, I have tested

hypotheses relating to mechanisms of wound-induced EGFR activation. The data presented in

Chapters Two through Six, combined with previously published reports from our and other

laboratories (Block and Klarlund, 2008; Block et al., 2004; Boucher et al., 2007; Mazie et al.,

2006; Xu et al., 2006; Yin et al., 2007; Zieske et al., 2000), have led to the development of a

working model for mechanisms of wound-induced EGFR activation (Figure 68). The proximal

mechanism of EGFR activation is the proteolytic release of membrane-bound ligands, which is

regulated by activation of Src-family kinases (SFKs). I have found that multiple pathways

converge on SFK signaling to regulate EGFR activation. One pathway is dependent on signaling

by extracellular ATP: wounding causes the release of ATP from cells near the wound edge via

cell rupture and unspecified channels; ATP then binds type-2 purinergic receptors (P2R) at the

cell surface, which signal to SFKs through the actions of phospholipase D2 (PLD2) and its

product, phosphatidic acid (PA). Since ATP is diffusible, ATP/PLD2 signaling can stimulate

EGFR activation at a distance from wounds. In a distinct pathway that functions solely in cells

near (<250 m from) the wound edge, an undefined wound sensor stimulates activation of the

97
focal adhesion kinase Pyk2. Activated Pyk2 then binds to and activates SFKs, leading to EGFR

activation. Finally, my data suggest the presence of a third distinct pathway, which triggers SFK

and EGFR activation in response to a physically unconstrained, or free edge.

Figure 68 A working model for mechanisms of EGFR activation after epithelial wounding

Evidence presented in this dissertation indicates that after epithelial wounding, multiple signaling pathways trigger

activation of the EGFR via SFK-mediated proteolytic ligand release. ATP (dark grey pathway) is released from

ruptured cells and unspecified membrane channels. Extracellular ATP binds type-2 purinergic receptors (P2R) and

stimulates SFK and EGFR activation through phospholipase D2 (PLD2) and phosphatidic acid (PA) signaling. Pyk2

(light grey pathway) is activated by undefined wound sensors and binds to and activates SFKs. Finally, the presence

of a free edge stimulates SFK activation through unspecified mechanisms.

All good working models stimulate more questions than they answer. In this chapter, I

will discuss the validity of this model: what are the shortcomings of this research that may have

confounded my interpretations? I will also discuss key areas for future research, specifically,

98
what is the molecular nature of the wound sensors that activate Pyk2 and that respond to a free

edge. Finally, I will discuss the implications of this model for wound healing in the clinic, and

for other pathologies that are affected by EGFR signaling.

7.2 SHORTCOMINGS

The major shortcoming of these studies leads us to question whether the model depicted in

Figure 63 is generally applicable to all epithelial wounds: my conclusions have been based

almost entirely on experiments with a human corneal epithelial cell line, and furthermore, I have

used unconventional wounding models. Are these conclusions relevant to actual wounds in any

epithelium?

First, the question of wounding models: in place of traditional scrape wounding, I have

used wounding models based on the removal of agarose barriers from the midst of otherwise

confluent epithelial sheets (Figures 5 and 6). By so doing, I have minimized the signaling input

from cell damage, which is not necessary for the induction of motility (Block et al., 2004;

Poujade et al., 2007) (although it may be necessary for other aspects of the normal healing

response). I therefore believe that our models allow the detection of signals that induce motility.

But are the signals I detect the same as those created by scrape-wounding? Using the

agarose methods, we have previously reported that the EGFR is activated after wounding

through the actions of HB-EGF (Block et al., 2004). Almost simultaneously, another report was

published that reached the same conclusions by using scrape-wounding (Xu et al., 2004). The

activation of Pyk2 after scrape-wounding has not been reported. Therefore, to increase

99
confidence that our agarose methods do not induce signaling unrelated to wounding, my studies

might have benefited from comparisons to scrape-wounding experiments.

Do the signals I detect after agarose removal have the same relative impact on EGFR

activation as they would after scrape-wounding? Scrape-wounding causes 5-10 fold higher

amounts of ATP to be released than does the removal of agarose barriers. Therefore,

extracellular ATP should have a much greater relative signaling role during scrape-wounding

studies. Elimination of ATP with apyrase did not alter healing rates, even after scrape-wounding

of corneas in organ culture (Figure 15C), so I believe that the increased signaling influence of

ATP during scrape-wounding experiments masks other signals that are necessary for EGFR

activation and healing.

In this dissertation, I have applied our wounding models almost exclusively to the study

of a human corneal epithelial cell line, HCLE cells. While these are a good model for native

corneal epithelial cells (Gipson et al., 2003), I have in two instances used additional cell types to

strengthen my conclusions: I used secondary cultures of rabbit corneal epithelial cells and rabbit

corneal organ cultures to increase confidence in my conclusion that extracellular ATP is not

necessary for wound healing (Figure 15), and I used MDCK cells to increase confidence in my

conclusion that free edges in epithelial sheets activate the EGFR (Figure 59). Some of the other

major findings of this dissertation (such as Figures 13, 22, 33, and 47) could have benefited from

confirmation in human neonatal foreskin epidermal keratinocytes (HEKn) or in MDCK cells.

Similar findings in these other cell types would raise confidence that the mechanisms of EGFR

activation shown in figure 68 are applicable to wounds in other tissues.

While it is clear that wounding activates the EGFR by proteolytic ligand release in vivo

(Shirakata et al., 2005; Tokumaru et al., 2000; Zieske et al., 2000), it is unclear whether the

100
upstream mechanisms I have identified operate in vivo as well. One would expect that

mechanisms of EGFR activation would vary in cell culture compared to living tissue and also

from one tissue to another, in large part due to highly variable physical and biochemical

microenvironments. For example, cultured cells are bathed in media that may be more

conducive to signaling by soluble factors than is the in vivo environment. In the same respect,

signaling by extracellular ATP may be less important in the skin than it is in the fluid

environment of the cornea. Detecting the release of ATP into the tear film or other wound fluids

should be feasible.

Whether signaling molecules such as ATP, PLD2, and Pyk2 regulate epithelial wound

healing in vivo could be tested using gene knockout strategies. Pyk2 knockout mice have been

bred (Okigaki et al., 2003) and live to adulthood. Wound healing in their corneas, skin, or any

epithelia could be compared to that in wild-type mice. Healing stimulated by EGF should also

be assessed to confirm that the loss of Pyk2 signaling creates no general defect in cell motility.

Live animal studies were beyond the scope of this research, but my findings in cultured cells

have generated testable hypotheses for future study.

A final minor experimental shortcoming also deserves mention. I have measured the

release of amphiregulin (AR) from HCLE cells by ELISA of tissue culture supernatants. As

discussed in Chapter Two, the amounts of AR detected by the ELISA are far too low, perhaps

100-fold too low, to induce measurable activation of the EGFR, although EGFR activation

clearly depends on AR (Figure 24). This implies that I am only detecting a fraction of the

bioactive AR that may or may not be representative of the entire bioactive pool. In experiments

where I have measured EGFR activation and AR release in parallel, AR quantities fluctuated

similarly with EGFR activation, suggesting that I am indeed measuring a representative fraction

101
of the active AR pool. On the other hand, AR is captured from the culture medium by the

EGFR, which is internalized after activation (Wiley, 2003), suggesting that AR levels detected

after wounding may over-estimate the degree of ligand release. I have attempted to address this

issue in two ways. First, I have tried to use high salt concentrations to remove bioactive AR

from the cell surface prior to analysis. Even after removal of the salt, which interfered with the

ELISA, I could not detect increased AR. A better approach would have been to measure AR

concentrations in the presence of an EGFR antibody that blocks ligand binding and thus greatly

inhibits capture from the culture medium (Oehrtman et al., 1998). Second, I have attempted to

measure AR release by Western Blotting with two different C-terminal-directed antibodies that

detect both the inactive (full-length) and active (cleaved) forms. Unfortunately, I was not able to

detect AR in HCLE cells by Western Blotting with either antibody.

7.3 WHAT ARE THE WOUND SENSORS?

In this dissertation, I have identified three signaling pathways that promote EGFR activation

after wounding (Figure 68): Signaling by extracellular ATP, by Pyk2, and by free edges. My

data indicate that these signals induce motility in wounded epithelia, but what are the molecular

signals, the wound sensors, that kick-start these three pathways? Some possibilities will be

discussed in the next sections.

102
7.3.1 Wound sensors in the extracellular ATP-dependent pathway

In the case of signaling by extracellular ATP, cell damage is clearly a wound sensor because cell

damage is responsible for the bulk of ATP released immediately after wounding. However, in

Chapter Four, I have shown that ATP, albeit at lower concentrations, is released independently

of cell damage. This implies that there is a second wound sensor that regulates ATP release

through an undefined channel. Experiments attempting to identify the ATP channel were

difficult to interpret, but the additional experiments suggested in Chapter Four could more

rigorously test the hypothesis that connexin hemichannels regulate ATP release. Release through

the undefined channel may be regulated by gene expression or membrane trafficking, or by

channel gating, either through mechanical factors such as stretch, or through cell signaling.

7.3.2 Wound sensors in the Pyk2 pathway

The identification of a Pyk2-dependent pathway for EGFR activation near the wound edge is

significant because it allows us to work upstream from Pyk2 to identify a wound sensor. Well-

characterized proximate causes for Pyk2 activation include signaling by integrins, by calcium,

and by reactive oxygen species (Avraham et al., 2000; Schlaepfer et al., 1999).

7.3.2.1 Integrin signaling

Integrins are heterodimeric adhesive receptors, comprised of and chains, that mediate cell-

cell and cell-extracellular matrix interactions. Twenty-four different receptors are formed by the

18 and 8 subunits described in mammals, and they are named according to subunit

composition (e.g. v1). Integrins are allosterically regulated by intracellular binding proteins

103
that prime the integrins for extracellular ligand binding (inside-out signaling) and by

extracellular ligands that induce intracellular signals (outside-in signaling) (Humphries et al.,

2003; Hynes, 2002; Luo et al., 2007; Shattil and Newman, 2004). Integrins also serve as a

transmembrane linkage between the extracellular matrix and the actin cytoskeleton (Blystone,

2004). A subset of / chain combinations have been described in the corneal epithelium,

primarily combinations that bind fibronectin, laminin, and collagen (Stepp, 2006).

A well-known intracellular signaling pathway stimulated by integrin adhesion is the

activation of Src-family kinases (Mitra and Schlaepfer, 2006; Playford and Schaller, 2004;

Reichelt, 2007), which has been reported to link adhesion to EGFR activation (Cabodi et al.,

2004; Kuwada and Li, 2000; Moro et al., 2002; Moro et al., 1998).

After wounding, integrin-based adhesions are formed rapidly at the edge of the wound

(Zaidel-Bar et al., 2003), but even cells in confluent monolayers exhibit basal cell motilities that

require replacing old integrin adhesions with new. It is therefore unclear whether a cell can

sense a wound based on integrin binding alone. Activation of FAK is an early signaling event

following integrin/matrix binding (Schlaepfer et al., 1999). I observed elevated FAK activities in

HCLE cells after attachment from suspension, but I did not observe FAK activation after

wounding (Figure 41A), suggesting that integrin/matrix binding is not a significant wound

sensor. Consistent with this conclusion, I have observed that treatments with integrin-blocking

antibodies and peptides did not inhibit wound-induced EGFR activation, even though they

blocked migration in wound healing assays.

7.3.2.2 Calcium signaling

Intracellular calcium concentrations are tightly regulated and many signaling molecules, such as

Pyk2, SFKs, and the EGFR can be activated by increased calcium concentration (Avraham et al.,
104
2000; Horiuchi et al., 2007; Schlaepfer et al., 1999; Thomas and Brugge, 1997). Minutes after

wounding, calcium increases have been observed in cells near the wound edge (Isakson et al.,

2001; Klepeis et al., 2001; Sammak et al., 1997), which coincides temporally and spatially with

the Pyk2-mediated activation of the EGFR. Calcium may therefore be a signaling intermediate

connecting wounding with Pyk2 and the EGFR. A number of future studies could explore this

possibility and eventually determine the proximate cause of increased calcium concentrations,

which could be a wound sensor.

To determine whether wound-induced Pyk2 activation occurs as a result of calcium

signaling, calcium signaling could be blocked by the intracellular calcium chelator BAPTA-AM.

Calcium activates Pyk2 by associating with calmodulin and binding near the Pyk2 N-terminus

(Kohno et al., 2008; Schaller, 2008). Therefore, as a complimentary approach, I could test

whether expression of the Pyk2 N-terminus (lacking the kinase domain) (Lipinski et al., 2006)

blocks endogenous Pyk2 activation by wounding. If BAPTA-AM and the Pyk2 N-terminus

block wound-induced Pyk2 activation, I would next determine whether the source of calcium is

intracellular stores (the ER) or extracellular fluid by wounding cells in the presence of the

extracellular calcium chelator EGTA or the Ca2+/ATPase inhibitor thapsigargen. I would use

this approach to examine wound-induced Pyk2 activation as well as changes in calcium

concentration measured by fluorometric dyes.

Studies in corneal, alveolar, and intestinal epithelial cells and bovine endothelial cells

have indicated that calcium signaling in cells near the wound edge is due to the influx of

extracellular calcium (Isakson et al., 2001; Klepeis et al., 2001; Rao et al., 2001; Sammak et al.,

1997). Furthermore, calcium influx is necessary for epithelial migration (Rao et al., 2001; Rao et

al., 2006). Calcium influx in cells near the wound edge was found to occur independently of cell

105
rupture, suggesting the presence of a wound-sensing calcium channel. Calcium channels can be

voltage-gated or non-voltage-gated, and calcium influx in non-excitable cells occurs primarily

through store-operated or stretch-activated channels, both of which have been shown to

contribute to cell motility (Duncan and Collison, 2002; Lee et al., 1999; Maroto et al., 2005; Rao

et al., 2006; Yamada et al.). Store-operated channels supply increased cytosolic calcium to

replenish ER stores following IP3 signaling. If inhibitors of PLC, which generates IP3, do not

inhibit calcium influx, store-operated channels are not likely responsible for calcium influx.

Using fluorometric calcium indicators, pharmacological and kinetic analyses of calcium influx

could implicate a particular type of channel that could be further targeted through siRNA or

dominant-negative approaches (Jaffe, 2007; Trollinger et al., 2002).

7.3.2.3 Reactive oxygen species

Like signaling by calcium, signaling by reactive oxygen species (ROS) also can activate Pyk2

and coincides temporally and spatially with wound-induced Pyk2 activation (Nikolic et al., 2006;

Schlaepfer et al., 1999). ROS is the general term given to molecules such as hydrogen peroxide

and superoxide that catalyze the reversible oxidation of cellular proteins. ROS can be generated

as a byproduct of aerobic respiration, by the NADPH oxidase complex, or by arachidonic acid

metabolism (Rhee, 2006; Rhee et al., 2005). ROS can be detected directly by fluorescent dyes

and indirectly by Western Blotting for reduced and oxidized forms of proteins. Active-site thiols

of protein tyrosine phosphatases are especially prone to oxidation and inactivation by ROS,

which couples cellular ROS production to tyrosine kinase signaling networks. ROS activate

Pyk2 by deactivating the phosphatase PTP-PEST (Lyons et al., 2001). Interestingly, ROS

generation by NADPH oxidase or arachidonic acid metabolism can be stimulated by the small

GTPase Rac1 (Gregg et al., 2003). Consequently, Rac1 can generate localized ROS signaling

106
and PTP-PEST inactivation that contributes to lamellipodia formation and cell migration (Wu et

al., 2005).

To determine whether ROS generation is the wound sensor for the Pyk2 pathway, I

would first apply various imaging and immunoblotting techniques to determine if ROS are

generated after wounding by our methods. For example, ROS can increase the fluorescence of

the cell-permeable dye, dichlorodihydrofluorescein diacetate, which could be visualized before

and after wounding. If ROS are generated, I would then wound cells in the presence of

antioxidants and measure Pyk2 activation. I could then proceed with various experiments to

determine the source of the ROS, and whether Rac1 is involved in their generation.

A major theoretical kink in the ROS as wound sensor hypothesis is that ROS can

induce activation of the EGFR and of SFKs, either directly or through phosphatase deactivation

(Nakashima et al., 2005; Peus et al., 1998; Zhang et al., 2001), which would seem to circumvent

the requirement for Pyk2 I have observed in Chapter Five. Still, pursuing this hypothesis may

uncover localized signaling mechanisms, as has been observed in lamellipodia (Wu et al., 2005),

that could explain this apparent contradiction.

7.3.3 The free edge sensor

In Chapter Six I presented data indicating that the presence of a free, physically unconstrained

edge stimulates activation of the EGFR, while a constrained edge does not. The proximate cause

for EGFR activation in response to a free edge is SFK-mediated ligand release, but the

mechanism of SFK activation and the molecular nature of the free edge sensor are unknown.

Because edges constrained by acellular barriers block activation, the free edge sensor is unlikely

to be the absence of a particular cell/cell interaction (Abercrombie, 1979; Carmona-Fontaine et

107
al., 2008; Desai et al., 2009; Lauffenburger et al., 1998). More likely, cells sense free edges due

to the lack of mechanical force opposing the free edge.

Biological systems are acutely sensitive to mechanical forces in large part due to the

tension cells maintain in the cytoskeleton to stabilize cell shape and subcellular organization

(Ingber, 2006). Mechanical stimuli alter conformation of numerous proteins (mechanosensors)

either directly or by challenging the rigidity of the cytoskeleton, and thereby stimulate signaling

cascades (del Rio et al., 2009; Friedland et al., 2009; Giannone and Sheetz, 2006; Orr et al.,

2006; Tamada et al., 2004). Cells can also sense mechanical signals based on the tension of the

plasma membrane, with low tension promoting both endocytosis and actin-based protrusions

(Sheetz and Dai, 1996). SFKs are activated by various mechanosensors and thus transduce

mechanical signals into cellular responses (Han et al., 2004; Ingley, 2008; Na et al., 2008).

To determine if a mechanosensor is responsible for free edge-stimulated EGFR

activation, a logical first approach would be to study the cytoskeleton at the free edge. I could

microscopically examine the actin and microtubule cytoskeletons in cells cultured on plastic

strips and compare it to control cultures. Since cytoskeletal rearrangements occur downstream of

EGFR signaling, I would expect to see them in the presence of the EGFR inhibitor AG 1478 if

they are involved with the free edge sensor. Actin-based protrusions such as filopodia can act

like sensory organelles (De Joussineau et al., 2003; Weber, 2006), raising the possibility that a

lack of physical constraints could cause actin-based protrusions to stimulate signaling.

Pharmacological inhibition of the actin and/or microtubule cytoskeletons can inhibit

mechanotransduction (Baker et al., 2004; Craig et al., 2008), so I would determine the effects of

drugs such as colchicine and cytochalasin D on free edge-induced EGFR activation. I could also

determine how free edges alter activities of proteins that control cytoskeletal rearrangements,

108
like the small GTPases Rho, Rac, and Cdc42 (Welch and Mullins, 2002). Identifying a single

mechanosensory molecule will be difficult, but if its expression or phosphorylation is altered by

free edges, I may be able to find reasonable candidates following genomic and proteomic

analyses.

7.3.4 Redundancy with a purpose

My data suggest the presence of three distinct mechanisms for wound-induced EGFR activation,

and each of these suggests its own set of hypothetical wound sensors. The possibilities for

wound sensors I have discussed (cell damage, ATP release, integrin signaling, calcium signaling,

ROS signaling, and cytoskeletal signaling) are not mutually exclusive. Instead, I assert that

many wound sensors contribute to EGFR activation, and that they can vary in importance

depending on the tissue and type of wound.

Because EGFR activation is so important for the induction of motility, many

mechanisms have evolved to ensure that the EGFR is activated following a variety of injuries.

Many of these mechanisms are present in the epithelium alone, but many more exist at the whole

tissue level. In the cornea for example, the EGFR is activated by EGF from tears, by EGF

secreted from infiltrating immune cells, and the EGFR is transactivated by HGF secreted from

keratocytes in the stroma (Klenkler and Sheardown, 2004; Klenkler et al., 2007; Spix et al.,

2007). Additionally, wounding may affect many of the systems that modulate EGFR avidity or

signaling, such as receptor trafficking (Curto et al., 2007; Gipson, 2004; Partridge et al., 2004;

Wiley, 2003), electrical currents (Zhao, 2009; Zhao et al., 1999), or EGFR oligomerization with

other signaling elements (Li et al., 2001; Warren and Landgraf, 2006; Yu et al., 2009).

109
 7.4 IMPLICATIONS FOR WOUND HEALING THERAPY

Even with all of the mechanisms in place to ensure robust activation and efficient signaling of

the EGFR after wounding, there are still many situations when wounds do not heal fast enough,

or at all. The broad goal of this dissertation was to identify molecular mechanisms of wound-

induced EGFR activation so that new therapeutic strategies may be developed for improving

normal and pathological healing.

The observation that stimulation of growth factor receptors, such as the EGFR, promote

epithelial restitution is not new, but growth factor therapies have met with limited clinical

success, primarily due to the poor stability of the proteins in the wound environment (Chen et al.,

1997; Hardwicke et al., 2008; Shell, 1982). HB-EGF and AR, ligands that have been found to

induce autocrine EGFR activation after wounding, adhere to cell surface glycosaminoglycans

after release, leading to a pool of active ligands that remains cell-associated (Harris et al., 2003).

Thus, unlike treatment with EGF that can act in hit and run fashion, treatment of wounded

epithelia with factors that promote autocrine ligand release leads to long-lasting and efficient

signaling (Shvartsman et al., 2002).

By investigating mechanisms of wound-induced EGFR activation, I have identified

signaling pathways that could be therapeutically stimulated to promote autocrine EGFR ligand

release. Treatment with the non-hydrolysable nucleotide analog ATPS or the water soluble

phosphatidic acid analog C8-PA warrant further study in animal models to determine if they can

induce long-lasting EGFR activation and healing. Additionally, factors known to stimulate Pyk2

or PLD2, such as acetylcholine or glial cell derived neurotropic factor (GDNF), might be

considered for future wound healing studies (Avraham et al., 2000; Exton, 2002).

110
 Recently, many advances in drug delivery and gene therapy have made promising strides

for wound healing therapies (Hardwicke et al., 2008; Mohan et al., 2005; Torchilin, 2006). The

corneal epithelium is an attractive model for gene therapy because it is easily accessed and

transgenic cells will eventually be lost due to natural sloughing. Other novel therapies include

proteins that are chemically modified or polymer-impregnated to achieve controlled release at

the wound site (Hardwicke et al., 2008; Hori et al., 2007). Similar controlled release techniques

with ATPS and C8-PA, and gene therapy techniques with PLD2 and Pyk2 should be examined

to test their safety and efficacy in promoting wound healing.

In addition to suggesting therapeutic targets for enhancing cell motility, the mechanistic

insights gained from my studies should help to understand how underlying pathologies disrupt

the normal healing process. For example, the loss of corneal nerves may impair healing due to

the loss of secreted factors such as GDNF and acetylcholine, which can stimulate PLD and/or

Pyk2 signaling (Cheng et al., 2003; Montiel et al., 2007; Nishida and Yanai, 2009; Sariola and

Saarma, 2003; You et al., 2001). Also, factors such as elevated tear osmolarity in dry eye and

high glucose in diabetes that affect EGFR signaling could be investigated for their affects on

PLD and Pyk2 signaling (Akhtar et al., 2009; Fischer et al., 2004; Lezama et al., 2005; Tomassen

et al., 2004; Xu et al., 2009).

Another clinical benefit of clarifying signals that control wound-induced EGFR

activation is the ability to identify risks of various ocular and systemic drugs. Cancer-fighting

anti-EGFR therapies cause persistent corneal epithelial and epidermal defects (Forster et al.,

2008; Garibaldi and Adler, 2007; Johnson et al., 2009), so drugs that interfere with signaling

upstream of the EGFR may have similar side-effects. Epithelial health, especially after

111
wounding, will need to be closely monitored in patients receiving new anti-cancer drugs that

target SFKs, Pyk2, and PLD2 (Bagi et al., 2008; Kim et al., 2009; Scott et al., 2009).

7.5 IMPLICATIONS FOR EGFR-RELATED PATHOLOGIES

The mechanisms I have identified in regulating EGFR activation (Figure 63) may not be limited

to wounded epithelia, but may also be relevant in other biological systems. I expect my findings

to be significant for pathologies that feature dysregulation of EGFR signaling, such as cancer and

cardiac hypertrophy.

7.5.1 Cancer

Many of the cellular processes that regulate epithelial migration after wounding also regulate the

motility of cancer cells (Friedl and Gilmour, 2009). EGFR activation is especially important for

cell migration during metastasis, the process by which cancer cells exit the primary tumor site,

enter the blood or lymphatic system, and take residence in other tissues (Xue et al., 2006;

Yamaguchi and Condeelis, 2007). As in wounds, the EGFR can be activated in cancer cells by

SFK-mediated proteolytic ligand release stimulated by GPCR agonists or other environmental

cues (Bhola and Grandis, 2008; Fischer et al., 2003; Higashiyama et al., 2008).

The intracellular signaling pathways I have identified as being mediators of EGFR ligand

release, PLD2 and Pyk2, have both been implicated in promoting cell migration and invasion in

a variety of cancers (Behmoaram et al., 2008; Foster and Xu, 2003; Huang and Frohman, 2007;

Iiizumi et al., 2008; Roelle et al., 2008; Scott et al., 2009; Sun et al., 2008; Zheng et al., 2006).

112
Despite the importance of EGFR signaling in cancer, only three studies I know of have

investigated whether either PLD or Pyk2 signaling can trigger EGFR activation (Andreev et al.,

2001; Dewar et al., 2007; Li and Malik, 2005). Their findings support my conclusion that PLD2

and Pyk2 promote cell motility through EGFR signaling. Thus, cancers over-expressing Pyk2 or

PLD2 may be sensitive to anti-EGFR therapies. Recent reports indicate that inhibiting signaling

pathways, such as SFKs and the EGFR, in combination is more effective than inhibiting either

alone (Egloff and Grandis, 2008). Therefore, inhibitors of PLD2 or Pyk2 should be tested in

combination with EGFR-blocking drugs.

7.5.2 Cardiac Health

Activation of the EGFR by proteolytic ligand release is important for the development and

function of the cardiovascular system (Booz and Baker, 1995; Eguchi et al., 2003; Iwamoto and

Mekada, 2006). In fact, most HB-EGF null mice die before weaning due to severe heart valve

defects (Iwamoto et al., 2003; Jackson et al., 2003). In the adult heart, EGFR ligand signaling is

important for cardiac remodeling, which is a constant dynamic process that responds to

hypertension, atherosclerosis, and heart attack. In cases of excessive cardiac remodeling, EGFR

signaling causes cardiac hypertrophy and fibrosis, leading to poor heart function and death. I

expect that my findings may be relevant to the study of cardiac hypertrophy.

A major regulator of myocardial EGFR activation, especially in response to hypertension,

is angiotensin II (Ang II). In this context, Ang II promotes cardiac hypertrophy through

transactivation of the EGFR by proteolytic ligand release (Chan et al., 2006; Higuchi et al., 2007;

Kagiyama et al., 2002; Smith et al., 2004; Thomas et al., 2002; Touyz and Berry, 2002), but the

intracellular signals that connect the Ang II receptor with EGFR ligand shedding are not clear.

113
Interestingly, Ang II stimulates PLD activity (Bollag et al., 1990) and inhibition of the PLD2

isoform by overexpression of a dominant-negative mutant inhibited Ang II-stimulated

hypertrophy in cultured vascular smooth muscle cells (Parmentier et al., 2006). Similarly, Ang II

signaling activates Pyk2, which contributes to the hypertrophic phenotype (Hirotani et al., 2004;

Menashi and Loftus, 2009; Touyz and Berry, 2002). Taken together with my findings (Figure

68), these results suggest the hypothesis that Ang II stimulates the EGFR through PLD2 and/or

Pyk2 signaling. Also, therapies that antagonize PLD2 or Pyk2 signaling may prevent cardiac

hypertrophy.

7.6 SUMMATION

The function of a wounded tissue and the health of an organism is jeopardized as long as an

epithelial wound remains open. This is especially true in the cornea, where even subtle

inflammation or scaring can impair vision. Cell migration is absolutely required for the closure

of epithelial wounds, and EGFR activation has been identified as an essential signal for the

induction of motility after wounding. Re-epithelialization rates determine clinical outcomes, yet

there are few effective therapies for enhancing cell migration. The broad goal of this dissertation

research was to clarify the cellular signals that stimulate EGFR activation after wounding so that

therapies may be developed to ameliorate normal and pathological healing.

To study EGFR activation, I have developed novel wounding models that minimize cell

damage and allow biochemical analysis of cells specifically near to, or far from, the wound edge.

I have found that multiple pathways converge on SFK signaling to regulate EGFR activation. In

cells near and far from wounds, extracellular ATP stimulates EGFR activation through PLD2

114
signaling. In distinct pathways that functions solely in cells near the wound edge, Pyk2 and an

undefined free edge sensor trigger SFK and EGFR activation. These findings lay the

groundwork for future studies to identify wound sensors and suggest new therapeutic targets for

modulating EGFR signaling and cell motility.

115
 8.0 MATERIALS AND METHODS

8.1 MATERIALS

Antibodies against phospholipase D2 were the generous gift of Dr. Sylvain G. Bourgoin

(Universit Laval, Quebec, QC, Canada). Antibodies (followed by catalog number) against

phospholipase D1 (sc-28314), EGFR (sc-03), EGFR phosphorylated on Tyr-1173 (sc-12351R),

ERK1 (sc-93), ERK1/2 phosphorylated on Tyr-204 (sc-7383), Pyk2 phosphorylated on Tyr-402

(sc-11767R), FAK phosphorylated on Tyr-397 (sc-11765R), and c-Src (mouse monoclonal used

for immunoprecipitation, sc-8056) were from Santa Cruz Biotechnology (Santa Cruz, CA);

antibodies against EGFR phosphorylated on Tyr-845 (2231), SFK phosphorylated on Tyr-419

(2101), SFK non-phosphorylated on Tyr-419 (2102), and c-Src (rabbit polyclonal used for

immunoblotting, 2109) were from Cell Signaling Technology (Danvers, MA); antibodies against

E-cadherin (610182), Pyk2 (610549), and Fyn (610163) were from BD Biosciences; antibodies

against MARCKS phosphorylated on serines 152 and 156 (442709) were from EMD Chemicals

(San Jose, CA); antibodies against paxillin and ZO-1 were from Invitrogen (Carlsbad, CA);

antibodies against -Actin (A5316) and FLAG were from Sigma-Aldrich (St. Louis, MO).

Antibodies against Ki67 (RM-9106) were from Thermo Fisher Scientific (Waltham, MA).

Neutralizing antibodies against AR, HB-EGF, and TGF were from R&D Systems

(Minneapolis, MN). Tyrphostin AG 1478, GM 6001, GM 6001 negative control, PP2, and Src

116
Kinase inhibitor-I were from EMD Chemicals (Gibbstown, NJ). ATP, grade VII apyrase from

potato, type V adenosine deaminase from bovine spleen, reactive blue 2, and polyHEMA were

from Sigma-Aldrich. Phosphatidic acid (1,2-dioctanoyl-sn-glycero-3-phosphate) and

phosphatidylinositol standards were from Avanti Polar Lipids (Alabaster, AL). Vybrant DiO,

DiD, and Alexa Fluor-conjugates were from Invitrogen. The peptides Gap-26

(VCYDKSFPISHVR) and Gap-27 (SRPTEKTIFII) were custom synthesized and provided at

70% purity by GenScript (Piscataway, NJ). Cell culture reagents were from Mediatech

(Manassas, VA) and other reagents and supplies were from ISC BioExpress (Kaysville, UT) or

Thermo Fisher, unless noted.

8.2 CELL CULTURE

Human corneal limbal epithelial (HCLE) cells have been immortalized by abrogation of

p16INK4A/Rb and p53 functions and overexpression of the catalytic subunit of the telomerase

holoenzyme (Gipson et al., 2003). HCLE cells were cultured in Keratinocyte Serum-Free

Medium (KSFM, Invitrogen) supplemented with 0.3 mM CaCl2, 25 g/ml bovine pituitary

extract, 0.1 ng/ml human recombinant EGF, 50 IU/ml penicillin, and 50 g/ml streptomycin. At

least 16 hours prior to stimulations, cells were washed and cultured in the same medium without

pituitary extract and EGF. HCLE monolayers were used for signaling studies, while stratified

cultures of HCLE cells were used for wound healing assays (see below). For some experiments,

I used secondary rabbit corneal epithelial cells. These cells were prepared from limbal epithelial

explants from rabbit eyes from Pel-Freez Biologicals (Rogers, AR) according to established

protocols (Ebato et al., 1987).

117
 8.3 WOUND HEALING ASSAY

A modification of the traditional scrape wounding assay was used as described previously (Block

et al., 2004). Molds for casting agarose strips were made by cutting 1 x 10 mm strips from

rectangles of vinyl vehicle lettering from the Letterz-Numberz Kit (Chroma International Inc.,

Alcoa, TN). Vinyl pieces were used right out of the package but were not sterilized. The

rectangles were attached to the growing surfaces of tissue culture plates by static cling and 2.5%

agarose with 0.6% glycerol was pipetted into the mold and allowed to dry overnight. The molds

were removed and cell suspensions added according to standard tissue culture procedures.

HCLE cells were grown to confluence around the agarose strips and were induced to

differentiate into a stratified epithelium (Gipson et al., 2003). Cells were transferred to

Dulbeccos Modified Eagle Medium:F-12 1:1 with 0.5%-2% newborn calf serum, the agarose

strip was removed, and healing was allowed to progress 14-24 hours before fixation or

preparation of cell extracts. Micrographs of cultures were obtained before and after healing and

wound areas were quantified using the region tracing utility in MetaMorph software (MDS

Analytical Technologies, Sunnyvale, CA). Experiments with mitomycin C have previously

demonstrated that wounds in HCLE cells heal as a result of cell migration.

8.4 WOUNDING MODEL FOR BIOCHEMICAL ANALYSIS OF SIGNALING IN

CELLS NEAR THE WOUND EDGE

A method for detecting signaling in cells 1-5 rows from the wound edge was used as described

previously (Block et al., 2004). A solution of 1% agarose was applied in a ring around the

118
periphery of tissue culture dishes to prevent cells from growing up the sides of the dishes. A

molten agarose solution (0.25% agarose with 0.6% glycerol) was then applied to the tissue dishes

on ice/water baths using an atomizer from Suave Naturals Ocean Breeze Spray Gel (Suave,

Chicago, IL) that generated a fine mist of agarose droplets. After drying the dishes at 65C for

30 minutes, cells were plated by standard procedures and were allowed to grow to confluence

around the droplets. To stimulate cells, fluid medium was replaced with semi-solid medium

containing 0.70% agarose. This top agarose layer was allowed to bond to the droplets and

solidify for 30 minutes in a 37C incubator followed by 15 minutes at room temperature. 37C

medium was added on top of the agarose and using a metal spatula, the top agarose (with

droplets attached) was gently lifted off as fluid medium flowed underneath. Stimulated cells

were then incubated for various times before biochemical analysis.

8.5 WOUNDING MODEL FOR BIOCHEMICAL ANALYSIS OF SIGNALING IN

CELLS FAR FROM THE WOUND EDGE

The HCLE cell populations were grown to confluence around a single agarose strip as described

for wound healing assay above. Cells were stimulated by the removal of this strip and reactions

were stopped 10 minutes later by plunging the bottom of the tissue culture dish into ice water

and replacing the medium with ice-cold PBS. Cells proximal to the wound were scraped and

discarded using a polyethylene cell lifter (Corning Inc., Corning, NY) trimmed to 7 mm,

effectively removing approximately 100 rows of cells from each side of the wound, and the

remaining cells were lysed directly in 1% SDS buffer.

119
 8.6 PRODUCTION OF POLYHEMA PLATES WITH PLASTIC STRIPS.

Tissue culture dishes (3.5 cm) were coated with 1 ml of 10% polyHEMA in 95% ethanol and

dried overnight at 37oC. The bottom parts of tissue culture dishes (Corning) were broken into

pieces and a 5% (weight/volume) solution in chloroform was prepared in 50 ml conical tubes on

an end-over-end rotator. Strips of the plastic solution were applied to the surface of the

polyHEMA with a narrow-bore 200 l pipette tip. Strips were roughly 0.5 mm wide, fitting 10-

20 rows of cells. As controls, the entire polyHEMA-coated surface was covered with a thin layer

of dissolved plastic. The plates were dried at 65oC overnight. HCLE cells were then seeded

according to standard procedures.

8.7 ANALYSIS OF SIGNALING AT VARIOUS DISTANCES FROM FREE EDGES

Populations of HCLE cells were seeded onto polyHEMA-coated tissue culture plates that were

half covered with dissolved plastic (with a single edge down the center). Reactions were stopped

by plunging the bottom of the tissue culture dish into ice water and replacing the medium with

ice-cold PBS. 0.5 cm-wide regions of sheets of cells were sequentially removed by scraping

with thin strips of autoradiography film, starting at the plastic/polyHEMA interface. A fresh 0.5

ml of ice-cold PBS was used to collect each subsequent region and corresponding regions from 3

dishes were pooled. Cells were collected by centrifugation before analysis.

120
 8.8 WESTERN BLOTTING

Following stimulations, cells were washed in ice-cold phosphate-buffered saline (171 mM NaCl,

10.1 mM Na2HPO4, 3.35 mM KCl, and 1.84 mM KH2PO4, pH 7.2) and lysed in SDS (1% in

H2O). The material was sonicated and protein contents were normalized after BCA assay

(Thermo Fisher). Lysates were loaded on SDS poly-acrylamide gels in the Mini-PROTEAN 3

gel apparatus (Bio-Rad, Hercules, CA). Immunoblotting was performed according to standard

procedure using optimized concentrations of antibodies. Multiple exposures of Western Blots

were collected, and densitometry of appropriate images was performed using Quantity One

software (Bio-Rad).

8.9 IMMUNOPRECIPITATION

Following treatment and washing in ice-cold PBS, cells were lysed in a non-denaturing lysis

buffer (50 mM Tris-Cl, 260 mM NaCl, 0.02% NaN3, 5 mM EDTA, 1% Triton X-100, 0.5 mM

Na3VO4, 50 mM NaF, 1 tablet/10 ml of protease inhibitor cocktail (Roche Diagnostics GmbH),

and 5 M pepstatin A, pH 7.5). Approximately 250 g of cellular protein was incubated with 30

l of protein-A sepharose slurry and 1 g of antibody on an end-over-end rotator at 4C

overnight. Bead pellets were washed twice with immunoprecipitation lysis buffer and three

times with lysis buffer made with 0.1% Triton X-100. SDS-PAGE sample buffer was added to

beads prior to Western Blotting.

121
 8.10 IMMUNOCYTOCHEMISTRY

Cells were fixed by adding 3.7% formaldehyde directly to the culture medium and incubating for

30 minutes. After three washes with PBS, the cells were permeabilized with 0.5% Triton X-100

in PBS with 10% calf serum (blocking buffer) for at least 1 hour. Subsequently, samples were

incubated overnight with appropriate dilutions of antibodies in blocking buffer. After washing,

samples were incubated with 1:500 dilutions of fluorophore-conjugated secondary antibodies for

1 hour. After three washes, coverslips were applied with 90% glycerol in PBS and 2.5% 1,4-

diazabicyclo-(2,2,2)-octane.

8.11 CONFOCAL FLUORESCENCE MICROSCOPY

Following immunocytochemistry, data were captured with an Olympus IX81 confocal

microscope equipped with a 60X oil objective (NA 1.4). Images were generated and analyzed

using Fluoview software (Olympus, Center Valley, PA).

8.12 EPIFLUORESCENCE MICROSCOPY

Following immunocytochemistry, data were captured with a Nikon TE2000E microscope

equipped with a 20X oil objective (NA 0.75). Images were generated and analyzed using

Elements software (Nikon Instrumets Inc., Melville, NY).

122
8.13 GENERATION OF IMAGES CONTAINING BOTH WOUND-PROXIMAL AND

WOUND-DISTAL CELLS

Following immunocytochemistry, seven contiguous aligned fields (a total of 5 mm from the

wound edge) per sample were captured on a Nikon TE2000E automated microscope with a 10X

objective (NA 0.3) using the MetaMorph scanslide utility. For quantitation of phospho-ERK

signal intensities, images were acquired at identical exposures and the average intensities of 250

m- or 62.5 m-wide regions were recorded for the entire 5 mm width of the image using

MetaMorph. Debris and other technical artifacts from sample preparation were excluded from

quantitation.

8.14 IMMUNOFLUORESCENCE MICROSCOPY OF CELLS ON PLASTIC STRIPS

OR NEXT TO AGAROSE DROPLETS

As rhodamine-dextran was serendipitously found to bind strongly to polyHEMA, polyHEMA-

coated plates were stained by incubation with 1 mg/ml rhodamine-dextran in PBS for 1 hour.

The plates were then washed 5 times with PBS and dried at room temperature. 10 l/ml

Vybrant DiO (Invitrogen) was added to the plastic solution before application as strips. Agarose

was conjugated to fluorescein by dissolving 8 mg agarose in 200 l DMSO at 100oC, and

reacting overnight with 4 mg fluorescein isothiocyanate (Sigma-Aldrich) at room temperature. It

was diluted to 1.6 ml in water, chilled to 0oC, and the solidified agarose washed extensively with

water. The conjugate was diluted to 0.25% before application to plastic-covered plates. The cells

were fixed with 3.7% formaldehyde in PBS and cell membranes were stained with 100 l/ml of

123
Vybrant DiD (Invitrogen) in PBS overnight. Images of the cultures were captured with an

Olympus IX81 confocal microscope equipped with a 60X oil objective (NA 1.4) and xy and xz

projections were generated and analyzed using Fluoview software (Olympus).

8.15 LIVE CELL MICROSCOPY AND DETERMINATION OF CELL VELOCITIES

HCLE cells were stratified around agarose strips in a 12-well dish and medium was changed to

1:1 DMEM:F-12 with 10% NCS 24 hours prior to experiment. Cells received no treatment or 30

U/ml apyrase and agarose strips were left in place or removed, such that N=3 for each condition.

Cell migration was monitored for the subsequent 24 hours with a Live Automated Cell Imager

(Schugar et al., 2009): cells were maintained in an environmentally controlled chamber (37oC

and 5% CO2) on a robotic stage and were visualized with an inverted Nikon Eclipse TE 2000 U

microscope equipped with a 10X objective and Photometric ES Cool Snap CCD camera. Time-

lapse images were created and cell velocities were calculated using MetaMorph by manually

tracking individual cells. In each well, four cells from each of two regions at the wound edge and

four cells from each of two regions 2 mm from the wound were analyzed. Cells were chosen

arbitrarily, prior to viewing the time-lapse movie. The analysis was blinded to apyrase treatment,

and in cells distal to the edge, the analysis was blinded to wounding. Due to errors in stage

movement and manual tracking, the velocity of a fixed object was determined to be 0.1 m/min

which was subtracted from calculated cell velocities.

124
 8.16 RABBIT AND RAT CORNEAL ORGAN CULTURE

To inflict wounds in rabbit eyes for organ culture, a 7.5 mm diameter mark was made with a

trephine, and the epithelium was removed by means of an Algerbrush II (Alger Equipment Co.,

Lago Vista, TX) and a sharp forceps. The wounded corneas were excised with a 2-3 mm scleral

rim and placed on hemispheric supports made from the round end of transfer pipets (Samco

Scientific Corp., San Fernando, CA) in 12-well dishes. The wound was briefly stained with 0.1%

fluorescein and photographed. They were subsequently incubated submerged at 37oC in Ham's

F12 Medium:Dulbecco's Modified Eagle's Medium (1:1) with 0.3% gentamycin for 60 hours. I

prefer a submerged culture technique because it allows continuous access of inhibitors. The

corneas were then stained with 1% alizarin red in phosphate buffered saline and photographed

again. The percentage of healing was calculated using MetaMorph software based on the areas

of the wounds before and after incubation. Whole rat eyes (Pel-Freez Biologicals) were

wounded by abrasion with an Algerbrush and cultured in the presence of no treatment or 30 U/ml

apyrase, and fixed after 24-48 hours in culture. Corneas were excised from globes and the

expression of Ki-67 was examined by immunohistochemistry of whole mounts.

8.17 ASSAY FOR PHOSPHOLIPASE D ACTIVITIES

Cells were labeled with 10-20 Ci [3H]myristic acid for 4 hours, stimulated, and 1-butanol added

to a final concentration of 0.5% for 5 minutes. Reactions were stopped and lipids were extracted

by the method of Bligh and Dyer (Bligh and Dyer, 1959). The dried samples were resuspended

125
in chloroform:methanol (2:1), and the phospholipids resolved on oxalate-treated silica gel G/UV

TLC plates (Whatman) by developing with CHCl3:methanol:glacial acetic acid (50:15:2). The

plates were air-dried, treated with EN3HANCE (PerkinElmer, Waltham, MA), and exposed to a

Kodak Biomax XAR film at 80C. For quantitation, phosphatidyl-butanol and

phosphatidylcholine spots were cut from the plates, and the radioactivity measured in a Beckman

LS3800 counter. Labeled phosphatidyl-butanol was identified by co-migration with authentic

phosphatidyl-butanol, by its absence without addition of 1-butanol, and by its increase upon

stimulation of cells with phorbol 12-myristate 13-acetate (PMA or TPA). Labeled

phosphatidylcholine was identified by co-migration with authentic phosphatidylcholine. The

amount of radioactivity in phosphatidyl-butanol was normalized to the amount in

phosphatidylcholine for each sample.

8.18 AMPHIREGULIN MEASUREMENT

Concurrent with wounding or hormonal stimulation, HCLE cells were cultured in freshly added

keratinocyte serum-free medium (KSFM) without EGF and pituitary extract for 10 minutes.

Conditioned medium was collected and subjected to centrifugation for 2 minutes at 5,000 x g.

The cell-free supernatants were aliquotted and stored at -20C until further processing.

Amphiregulin was measured in the supernatants using the DuoSet ELISA (R&D Systems)

according to manufacturers protocol. For normalization purposes, protein content of whole cell

extracts was determined by the BCA protein assay (Thermo Fisher).

126
 8.19 ATP MEASUREMENT

Concurrent with wounding or hormonal stimulation, HCLE cells were cultured in freshly added

KSFM without EGF and pituitary extract for 10 minutes. Conditioned medium was collected

and subjected to centrifugation for 2 minutes at 5,000 x g. The cell-free supernatants were

aliquotted and stored at -20C until further processing. ATP was measured in the cell-free tissue

culture supernatants with the ATP Bioluminescent Assay Kit (Sigma-Aldrich). Independent

standard curves were created for the drug-treated groups to control for effects of the drug or

altered pH on the assay. For normalization purposes, protein content of whole cell extracts was

determined by the BCA protein assay (Thermo Fisher).

To detect ATP degradation in situ, 10 Ci [-32P]-ATP was added to the cultures and an

aliquot was collected after 5 minutes for analysis by thin-layer chromatography as described

(Hagmann et al., 1999). ATP and ADP were excised from the thin-layer chromatography plate,

subjected to scintillation counting, and normalized to total radioactivity in the sample.

8.20 LUCIFER YELLOW AND RHODAMINE-DEXTRAN UPTAKE

Lucifer yellow and rhodamine-conjugated dextran were diluted to 0.5% in culture medium and

added to live HCLE cells for five minutes with desired treatments. After removal of dyes, cells

were washed three times with PBS and fixed with 3.7% formaldehyde in PBS before

fluorescence microscopy.

127
 8.21 SIRNA TRANSFECTIONS

SiRNA was transfected into subconfluent cells at the time of seeding using the siPORTTM

NeoFXTM lipid-based reverse transfection reagent (Applied Biosystems, Foster City, CA). Two

days after transfections, cells were re-seeded to generate confluent cultures and were used the

following day. For PLD experiments, the following oligonucleotides were designed according to

published sequences and standard rules (Pei and Tuschl, 2006) and were synthesized by Applied

Biosystems (sense strand): PLD1, GUUAAGAGGAAAUUCAAGCdTdT; PLD2 sequence a,

UGGGGCAGGUUACUUUGCUdTdT; PLD2 sequence b,

AGUCUUGAUGAGGUCUGCUCdTdT. For Pyk2 studies, multiple siRNA oligonucleotides

were used: CACAUGAAGUCCGAUGAGAdTdT (Sigma-Aldrich) and a Pyk2 SMARTpool that

contains at least four duplexes of undisclosed sequence (Millipore, Billerica, MA). As a control

for Pyk2 siRNA, I used MISSION siRNA Universal Negative Control #1 (Sigma-Aldrich).

BLAST searches (Altschul et al., 1990) with siRNA sequences revealed significant sequence

homologies with only the targeted mRNAs. For each experiment, the expression of target

protein was monitored by Western Blotting.

8.22 ADENOVIRAL INFECTIONS

FLAG-tagged Pyk2 and FLAG-tagged PRNK cloned into adenovirus vectors were obtained from

Dr. Joseph C. Loftus (Mayo Clinic, Scottsdale, AZ). Adenovirus encoding the tet-OFF

transactivator (used as adenovirus control) was from Dr. Ora A. Weisz (University of Pittsburgh,

Pittsburgh, PA) and adenovirus encoding GFP was from Dr. Michael P. Czech (University of

128
Massachusetts, Worcester, MA). Preliminary experiments using GFP-encoding virus helped to

determine the multiplicity of infection necessary for achieving expression in ~100% of cells. For

signaling studies, cells were infected at a multiplicity of 10 for 1 hr and were used the following

day. For wound healing studies, cells were infected at a multiplicity of 20 in a small volume (1 x

105 cells/l) for 1 hr, with gentle agitation every 15 min. After washing and centrifugation, cells

were seeded for wound healing assay and were used 2 days later.

8.23 STATISTICAL ANALYSIS

All experiments reported were repeated a minimum of three times with similar results. For some

figures, representative Western Blots are shown below densitometry from at least four replicates.

Quantitative data were plotted (means +/- S.D.) and analyzed by t-test or one-way ANOVA

followed by Bonferronis multiple comparisons test using Prism (GraphPad Software, La Jolla,

CA).

129
 APPENDIX A

LIST OF PEER-REVIEWED PUBLICATIONS CO-AUTHORED BY ETHAN R. BLOCK

Block,E.R, Tolino, M.A., and Klarlund, J.K. (2010). Pyk2 Activation triggers EGF receptor

signaling and cell motility after wounding sheets of epithelial cells. Submitted.

Block, E.R., Tolino, M.A., Lathrop, K.L., Sullenberger, R.S., Mazie, A.R., and Klarlund, J.K.

(2010). Edges in epithelial cell sheets stimulate epidermal growth factor receptor signaling.

Submitted.

Block, E. R., and Klarlund, J. K. (2008). Wounding sheets of epithelial cells activates the

epidermal growth factor receptor through distinct short- and long-range mechanisms. Mol Biol

Cell 19, 4909-17.

Spix, J. K., Chay, E. Y., Block, E. R., and Klarlund, J. K. (2007). Hepatocyte growth factor

induces epithelial cell motility through transactivation of the epidermal growth factor receptor.

Exp Cell Res 313, 3319-25.

130
Mazie, A. R., Spix, J. K., Block, E. R., Achebe, H. B., and Klarlund, J. K. (2006). Epithelial cell

motility is triggered by activation of the EGF receptor through phosphatidic acid signaling. J

Cell Sci 119, 1645-54.

Block, E. R., Matela, A. R., SundarRaj, N., Iszkula, E. R., and Klarlund, J. K. (2004). Wounding

induces motility in sheets of corneal epithelial cells through loss of spatial constraints: role of

heparin-binding epidermal growth factor-like growth factor signaling. J Biol Chem 279, 24307-

12.

131
 BIBLIOGRAPHY

Abbaci, M., Barberi-Heyob, M., Blondel, W., Guillemin, F. and Didelon, J. (2008).
Advantages and limitations of commonly used methods to assay the molecular permeability of
gap junctional intercellular communication. Biotechniques 45, 33-52, 56-62.

Abbracchio, M. P., Burnstock, G., Boeynaems, J. M., Barnard, E. A., Boyer, J. L.,
Kennedy, C., Knight, G. E., Fumagalli, M., Gachet, C., Jacobson, K. A. et al. (2006).
International Union of Pharmacology LVIII: update on the P2Y G protein-coupled nucleotide
receptors: from molecular mechanisms and pathophysiology to therapy. Pharmacol Rev 58, 281-
341.

Abercrombie, M. (1979). Contact inhibition and malignancy. Nature 281, 259-62.

Abram, C. L. and Courtneidge, S. A. (2000). Src family tyrosine kinases and growth
factor signaling. Exp Cell Res 254, 1-13.

Aderem, A. (1992). The MARCKS brothers: a family of protein kinase C substrates. Cell
71, 713-6.

Ahn, B. H., Kim, S. Y., Kim, E. H., Choi, K. S., Kwon, T. K., Lee, Y. H., Chang, J.
S., Kim, M. S., Jo, Y. H. and Min, D. S. (2003). Transmodulation between phospholipase D
and c-Src enhances cell proliferation. Mol Cell Biol 23, 3103-15.

Akhtar, S., Almubrad, T., Bron, A. J., Yousif, M. H., Benter, I. F. and Akhtar, S.
(2009). Role of epidermal growth factor receptor (EGFR) in corneal remodelling in diabetes.
Acta Ophthalmol 87, 881-9.

Ali, S. H. and DeCaprio, J. A. (2001). Cellular transformation by SV40 large T antigen:

interaction with host proteins. Semin Cancer Biol 11, 15-23.

Altschul, S. F., Gish, W., Miller, W., Myers, E. W. and Lipman, D. J. (1990). Basic
local alignment search tool. J Mol Biol 215, 403-10.

Andreev, J., Galisteo, M. L., Kranenburg, O., Logan, S. K., Chiu, E. S., Okigaki, M.,
Cary, L. A., Moolenaar, W. H. and Schlessinger, J. (2001). Src and Pyk2 mediate G-protein-
coupled receptor activation of epidermal growth factor receptor (EGFR) but are not required for
coupling to the mitogen-activated protein (MAP) kinase signaling cascade. J Biol Chem 276,
20130-5.

132
 Assoian, R. K. and Klein, E. A. (2008). Growth control by intracellular tension and
extracellular stiffness. Trends Cell Biol 18, 347-52.

Avraham, H., Park, S. Y., Schinkmann, K. and Avraham, S. (2000). RAFTK/Pyk2-

mediated cellular signalling. Cell Signal 12, 123-33.

Bagi, C. M., Roberts, G. W. and Andresen, C. J. (2008). Dual focal adhesion

kinase/Pyk2 inhibitor has positive effects on bone tumors: implications for bone metastases.
Cancer 112, 2313-21.

Bain, J., Plater, L., Elliott, M., Shpiro, N., Hastie, C. J., McLauchlan, H., Klevernic,
I., Arthur, J. S., Alessi, D. R. and Cohen, P. (2007). The selectivity of protein kinase
inhibitors: a further update. Biochem J 408, 297-315.

Baker, L. C., Wolk, R., Choi, B. R., Watkins, S., Plan, P., Shah, A. and Salama, G.
(2004). Effects of mechanical uncouplers, diacetyl monoxime, and cytochalasin-D on the
electrophysiology of perfused mouse hearts. Am J Physiol Heart Circ Physiol 287, H1771-9.

Baldwin, H. C. and Marshall, J. (2002). Growth factors in corneal wound healing

following refractive surgery: A review. Acta Ophthalmol Scand 80, 238-47.

Barrandon, Y. (2007). Crossing boundaries: stem cells, holoclones, and the

fundamentals of squamous epithelial renewal. Cornea 26, S10-2.

Bazan, H. E. (2005). Cellular and molecular events in corneal wound healing:

significance of lipid signalling. Exp Eye Res 80, 453-63.

Behmoaram, E., Bijian, K., Jie, S., Xu, Y., Darnel, A., Bismar, T. A. and Alaoui-
Jamali, M. A. (2008). Focal adhesion kinase-related proline-rich tyrosine kinase 2 and focal
adhesion kinase are co-overexpressed in early-stage and invasive ErbB-2-positive breast cancer
and cooperate for breast cancer cell tumorigenesis and invasiveness. Am J Pathol 173, 1540-50.

Beigi, R., Kobatake, E., Aizawa, M. and Dubyak, G. R. (1999). Detection of local
ATP release from activated platelets using cell surface-attached firefly luciferase. Am J Physiol
276, C267-78.

Belliveau, D. J., Bani-Yaghoub, M., McGirr, B., Naus, C. C. and Rushlow, W. J.

(2006). Enhanced neurite outgrowth in PC12 cells mediated by connexin hemichannels and ATP.
J Biol Chem 281, 20920-31.

Belmonte, C. (2007). Eye dryness sensations after refractive surgery: impaired tear
secretion or "phantom" cornea? J Refract Surg 23, 598-602.

Bement, W. M., Yu, H. Y., Burkel, B. M., Vaughan, E. M. and Clark, A. G. (2007).
Rehabilitation and the single cell. Curr Opin Cell Biol 19, 95-100.

Bhola, N. E. and Grandis, J. R. (2008). Crosstalk between G-protein-coupled receptors

and epidermal growth factor receptor in cancer. Front Biosci 13, 1857-65.

133
 Bindschadler, M. and McGrath, J. L. (2007). Sheet migration by wounded monolayers
as an emergent property of single-cell dynamics. J Cell Sci 120, 876-84.

Bligh, E. G. and Dyer, W. J. (1959). A rapid method of total lipid extraction and
purification. Can J Biochem Physiol 37, 911-7.

Blobel, C. P. (2005). ADAMs: key components in EGFR signalling and development.

Nat Rev Mol Cell Biol 6, 32-43.

Block, E. R. and Klarlund, J. K. (2008). Wounding sheets of epithelial cells activates

the epidermal growth factor receptor through distinct short- and long-range mechanisms. Mol
Biol Cell 19, 4909-17.

Block, E. R., Matela, A. R., SundarRaj, N., Iszkula, E. R. and Klarlund, J. K. (2004).
Wounding induces motility in sheets of corneal epithelial cells through loss of spatial constraints:
role of heparin-binding epidermal growth factor-like growth factor signaling. J Biol Chem 279,
24307-12.

Blystone, S. D. (2004). Integrating an integrin: a direct route to actin. Biochim Biophys

Acta 1692, 47-54.

Boitano, S. and Evans, W. H. (2000). Connexin mimetic peptides reversibly inhibit

Ca(2+) signaling through gap junctions in airway cells. Am J Physiol Lung Cell Mol Physiol 279,
L623-30.

Bollag, W. B., Barrett, P. Q., Isales, C. M., Liscovitch, M. and Rasmussen, H. (1990).
A potential role for phospholipase-D in the angiotensin-II-induced stimulation of aldosterone
secretion from bovine adrenal glomerulosa cells. Endocrinology 127, 1436-43.

Bonini, S., Rama, P., Olzi, D. and Lambiase, A. (2003). Neurotrophic keratitis. Eye 17,
989-95.

Booz, G. W. and Baker, K. M. (1995). Molecular signalling mechanisms controlling

growth and function of cardiac fibroblasts. Cardiovasc Res 30, 537-43.

Boucher, I., Yang, L., Mayo, C., Klepeis, V. and Trinkaus-Randall, V. (2007). Injury
and nucleotides induce phosphorylation of epidermal growth factor receptor: MMP and HB-EGF
dependent pathway. Exp Eye Res 85, 130-41.

Bours, M. J., Swennen, E. L., Di Virgilio, F., Cronstein, B. N. and Dagnelie, P. C.

(2006). Adenosine 5'-triphosphate and adenosine as endogenous signaling molecules in
immunity and inflammation. Pharmacol Ther 112, 358-404.

Braun, M., Lelieur, K. and Kietzmann, M. (2006). Purinergic substances promote

murine keratinocyte proliferation and enhance impaired wound healing in mice. Wound Repair
Regen 14, 152-61.

134
 Brown, D. C., Theaker, J. M., Banks, P. M., Gatter, K. C. and Mason, D. Y. (2002).
Cytokeratin expression in smooth muscle and smooth muscle tumours. Histopathology 41, 85-
94.

Bruzzone, R., Barbe, M. T., Jakob, N. J. and Monyer, H. (2005). Pharmacological

properties of homomeric and heteromeric pannexin hemichannels expressed in Xenopus oocytes.
J Neurochem 92, 1033-43.

Burdick, A. D., Ivnitski-Steele, I. D., Lauer, F. T. and Burchiel, S. W. (2006). PYK2

mediates anti-apoptotic AKT signaling in response to benzo[a]pyrene diol epoxide in mammary
epithelial cells. Carcinogenesis 27, 2331-40.

Burnstock, G. (2006). Purinergic signalling--an overview. Novartis Found Symp 276,

26-48; discussion 48-57, 275-81.

Burnstock, G. (2007). Physiology and pathophysiology of purinergic neurotransmission.

Physiol Rev 87, 659-797.

Cabodi, S., Moro, L., Bergatto, E., Boeri Erba, E., Di Stefano, P., Turco, E., Tarone,
G. and Defilippi, P. (2004). Integrin regulation of epidermal growth factor (EGF) receptor and
of EGF-dependent responses. Biochem Soc Trans 32, 438-42.

Cameron, J. D. (2007). Surgical and Nonsurgical Trauma. In Duane's Ophthalmology,

(ed.s W. T. a. E. A. Jaeger): Lipincott, Williams, and Wilkins.

Carmona-Fontaine, C., Matthews, H. K., Kuriyama, S., Moreno, M., Dunn, G. A.,
Parsons, M., Stern, C. D. and Mayor, R. (2008). Contact inhibition of locomotion in vivo
controls neural crest directional migration. Nature 456, 957-61.

Cazzolli, R., Shemon, A. N., Fang, M. Q. and Hughes, W. E. (2006). Phospholipid

signalling through phospholipase D and phosphatidic acid. IUBMB Life 58, 457-61.

Chan, H. W., Jenkins, A., Pipolo, L., Hannan, R. D., Thomas, W. G. and Smith, N.
J. (2006). Effect of dominant-negative epidermal growth factor receptors on cardiomyocyte
hypertrophy. J Recept Signal Transduct Res 26, 659-77.

Chen, P., Murphy-Ullrich, J. E. and Wells, A. (1996). A role for gelsolin in actuating
epidermal growth factor receptor-mediated cell motility. J Cell Biol 134, 689-98.

Chen, S. M., Ward, S. I., Olutoye, O. O., Diegelmann, R. F. and Kelman Cohen, I.
(1997). Ability of chronic wound fluids to degrade peptide growth factors is associated with
increased levels of elastase activity and diminished levels of proteinase inhibitors. Wound Repair
Regen 5, 23-32.

Chen, Y., Corriden, R., Inoue, Y., Yip, L., Hashiguchi, N., Zinkernagel, A., Nizet,
V., Insel, P. A. and Junger, W. G. (2006). ATP release guides neutrophil chemotaxis via P2Y2
and A3 receptors. Science 314, 1792-5.

135
 Chen, Z., Evans, W. H., Pflugfelder, S. C. and Li, D. Q. (2006). Gap junction protein
connexin 43 serves as a negative marker for a stem cell-containing population of human limbal
epithelial cells. Stem Cells 24, 1265-73.

Cheng, K., Zimniak, P. and Raufman, J. P. (2003). Transactivation of the epidermal

growth factor receptor mediates cholinergic agonist-induced proliferation of H508 human colon
cancer cells. Cancer Res 63, 6744-50.

Chikama, T., Wakuta, M., Liu, Y. and Nishida, T. (2007). Deviated mechanism of
wound healing in diabetic corneas. Cornea 26, S75-81.

Chu, E. K., Foley, J. S., Cheng, J., Patel, A. S., Drazen, J. M. and Tschumperlin, D.
J. (2005). Bronchial epithelial compression regulates epidermal growth factor receptor family
ligand expression in an autocrine manner. Am J Respir Cell Mol Biol 32, 373-80.

Cole, J., Tsou, R., Wallace, K., Gibran, N. and Isik, F. (2001). Early gene expression
profile of human skin to injury using high-density cDNA microarrays. Wound Repair Regen 9,
360-70.

Cotrina, M. L., Lin, J. H., Alves-Rodrigues, A., Liu, S., Li, J., Azmi-Ghadimi, H.,
Kang, J., Naus, C. C. and Nedergaard, M. (1998). Connexins regulate calcium signaling by
controlling ATP release. Proc Natl Acad Sci U S A 95, 15735-40.

Craig, D. H., Owen, C. R., Conway, W. C., Walsh, M. F., Downey, C. and Basson,
M. D. (2008). Colchicine inhibits pressure-induced tumor cell implantation within surgical
wounds and enhances tumor-free survival in mice. J Clin Invest 118, 3170-80.

Curto, M., Cole, B. K., Lallemand, D., Liu, C. H. and McClatchey, A. I. (2007).
Contact-dependent inhibition of EGFR signaling by Nf2/Merlin. J Cell Biol 177, 893-903.

D'Hondt, C., Ponsaerts, R., De Smedt, H., Bultynck, G. and Himpens, B. (2009).
Pannexins, distant relatives of the connexin family with specific cellular functions? Bioessays 31,
953-74.

Daniels, J. T., Harris, A. R. and Mason, C. (2006). Corneal epithelial stem cells in
health and disease. Stem Cell Rev 2, 247-54.

Danjo, Y. and Gipson, I. K. (1998). Actin 'purse string' filaments are anchored by E-
cadherin-mediated adherens junctions at the leading edge of the epithelial wound, providing
coordinated cell movement. J Cell Sci 111 (Pt 22), 3323-32.

Das, S., Langenbucher, A. and Seitz, B. (2005). Delayed healing of corneal epithelium
after phototherapeutic keratectomy for lattice dystrophy. Cornea 24, 283-7.

Daub, H., Weiss, F. U., Wallasch, C. and Ullrich, A. (1996). Role of transactivation of
the EGF receptor in signalling by G-protein-coupled receptors. Nature 379, 557-60.

136
 Davies, D. E., Polosa, R., Puddicombe, S. M., Richter, A. and Holgate, S. T. (1999).
The epidermal growth factor receptor and its ligand family: their potential role in repair and
remodelling in asthma. Allergy 54, 771-83.

Davis, E. A. and Dohlman, C. H. (2001). Neurotrophic keratitis. Int Ophthalmol Clin

41, 1-11.

de Amicis, F., Lanzino, M., Kisslinger, A., Cali, G., Chieffi, P., Ando, S., Mancini, F.
P. and Tramontano, D. (2006). Loss of proline-rich tyrosine kinase 2 function induces
spreading and motility of epithelial prostate cells. J Cell Physiol 209, 74-80.

De Joussineau, C., Soule, J., Martin, M., Anguille, C., Montcourrier, P. and
Alexandre, D. (2003). Delta-promoted filopodia mediate long-range lateral inhibition in
Drosophila. Nature 426, 555-9.

del Rio, A., Perez-Jimenez, R., Liu, R., Roca-Cusachs, P., Fernandez, J. M. and
Sheetz, M. P. (2009). Stretching single talin rod molecules activates vinculin binding. Science
323, 638-41.

DeMali, K. A. and Burridge, K. (2003). Coupling membrane protrusion and cell

adhesion. J Cell Sci 116, 2389-97.

Desai, R. A., Gao, L., Raghavan, S., Liu, W. F. and Chen, C. S. (2009). Cell polarity
triggered by cell-cell adhesion via E-cadherin. J Cell Sci 122, 905-11.

Dewar, B. J., Gardner, O. S., Chen, C. S., Earp, H. S., Samet, J. M. and Graves, L.
M. (2007). Capacitative calcium entry contributes to the differential transactivation of the
epidermal growth factor receptor in response to thiazolidinediones. Mol Pharmacol 72, 1146-56.

Dieckgraefe, B. K., Weems, D. M., Santoro, S. A. and Alpers, D. H. (1997). ERK and
p38 MAP kinase pathways are mediators of intestinal epithelial wound-induced signal
transduction. Biochem Biophys Res Commun 233, 389-94.

Dignass, A. U., Becker, A., Spiegler, S. and Goebell, H. (1998). Adenine nucleotides
modulate epithelial wound healing in vitro. Eur J Clin Invest 28, 554-61.

Dikic, I., Tokiwa, G., Lev, S., Courtneidge, S. A. and Schlessinger, J. (1996). A role
for Pyk2 and Src in linking G-protein-coupled receptors with MAP kinase activation. Nature
383, 547-50.

Dise, R. S., Frey, M. R., Whitehead, R. H. and Polk, D. B. (2008). Epidermal growth
factor stimulates Rac activation through Src and phosphatidylinositol 3-kinase to promote
colonic epithelial cell migration. Am J Physiol Gastrointest Liver Physiol 294, G276-85.

Duane, T. D., Tasman, W. and Jaeger, E. A. (2002). Duane's clinical ophthalmology

on CD-ROM, pp. 1 computer optical disc. [Philadelphia]: Lippincott Williams & Wilkins.

137
 Duncan, G. and Collison, D. J. (2002). Calcium signalling in ocular tissues: functional
activity of G-protein and tyrosine-kinase coupled receptors. Exp Eye Res 75, 377-89.

Duong, L. T., Lakkakorpi, P. T., Nakamura, I., Machwate, M., Nagy, R. M. and
Rodan, G. A. (1998). PYK2 in osteoclasts is an adhesion kinase, localized in the sealing zone,
activated by ligation of alpha(v)beta3 integrin, and phosphorylated by src kinase. J Clin Invest
102, 881-92.

Dupps, W. J., Jr. and Wilson, S. E. (2006). Biomechanics and wound healing in the
cornea. Exp Eye Res 83, 709-20.

Ebato, B., Friend, J. and Thoft, R. A. (1987). Comparison of central and peripheral
human corneal epithelium in tissue culture. Invest Ophthalmol Vis Sci 28, 1450-6.

Egloff, A. M. and Grandis, J. R. (2008). Targeting epidermal growth factor receptor

and SRC pathways in head and neck cancer. Semin Oncol 35, 286-97.

Eguchi, S., Frank, G. D., Mifune, M. and Inagami, T. (2003). Metalloprotease-

dependent ErbB ligand shedding in mediating EGFR transactivation and vascular remodelling.
Biochem Soc Trans 31, 1198-202.

el-Moatassim, C. and Dubyak, G. R. (1992). A novel pathway for the activation of

phospholipase D by P2z purinergic receptors in BAC1.2F5 macrophages. J Biol Chem 267,
23664-73.

Eskandari, S., Zampighi, G. A., Leung, D. W., Wright, E. M. and Loo, D. D. (2002).
Inhibition of gap junction hemichannels by chloride channel blockers. J Membr Biol 185, 93-
102.

Evans, W. H. and Boitano, S. (2001). Connexin mimetic peptides: specific inhibitors of

gap-junctional intercellular communication. Biochem Soc Trans 29, 606-12.

Evans, W. H., De Vuyst, E. and Leybaert, L. (2006). The gap junction cellular internet:
connexin hemichannels enter the signalling limelight. Biochem J 397, 1-14.

Evans, W. H. and Martin, P. E. (2002). Gap junctions: structure and function (Review).
Mol Membr Biol 19, 121-36.

Exton, J. H. (2002). Phospholipase D-structure, regulation and function. Rev Physiol

Biochem Pharmacol 144, 1-94.

Fang, Y., Vilella-Bach, M., Bachmann, R., Flanigan, A. and Chen, J. (2001).
Phosphatidic acid-mediated mitogenic activation of mTOR signaling. Science 294, 1942-5.

Farooqui, R. and Fenteany, G. (2005). Multiple rows of cells behind an epithelial

wound edge extend cryptic lamellipodia to collectively drive cell-sheet movement. J Cell Sci
118, 51-63.

138
 Fenteany, G., Janmey, P. A. and Stossel, T. P. (2000). Signaling pathways and cell
mechanics involved in wound closure by epithelial cell sheets. Curr Biol 10, 831-8.

Fini, M. E. and Stramer, B. M. (2005). How the cornea heals: cornea-specific repair
mechanisms affecting surgical outcomes. Cornea 24, S2-S11.

Fischer, O. M., Hart, S., Gschwind, A., Prenzel, N. and Ullrich, A. (2004). Oxidative
and osmotic stress signaling in tumor cells is mediated by ADAM proteases and heparin-binding
epidermal growth factor. Mol Cell Biol 24, 5172-83.

Fischer, O. M., Hart, S., Gschwind, A. and Ullrich, A. (2003). EGFR signal
transactivation in cancer cells. Biochem Soc Trans 31, 1203-8.

Fitsialos, G., Chassot, A. A., Turchi, L., Dayem, M. A., LeBrigand, K., Moreilhon,
C., Meneguzzi, G., Busca, R., Mari, B., Barbry, P. et al. (2007). Transcriptional signature of
epidermal keratinocytes subjected to in vitro scratch wounding reveals selective roles for
ERK1/2, p38, and phosphatidylinositol 3-kinase signaling pathways. J Biol Chem 282, 15090-
102.

Forster, C. G., Cursiefen, C. and Kruse, F. E. (2008). [Topical application of EGF for
the therapy of persisting corneal erosion under cetuximab treatment]. Ophthalmologe 105, 269-
73.

Foster, D. A. and Xu, L. (2003). Phospholipase D in cell proliferation and cancer. Mol
Cancer Res 1, 789-800.

Frame, M. K. and de Feijter, A. W. (1997). Propagation of mechanically induced

intercellular calcium waves via gap junctions and ATP receptors in rat liver epithelial cells. Exp
Cell Res 230, 197-207.

Frank, C., Keilhack, H., Opitz, F., Zschornig, O. and Bohmer, F. D. (1999). Binding
of phosphatidic acid to the protein-tyrosine phosphatase SHP-1 as a basis for activity
modulation. Biochemistry 38, 11993-2002.

Friedl, P. (2004). Prespecification and plasticity: shifting mechanisms of cell migration.

Curr Opin Cell Biol 16, 14-23.

Friedl, P. and Gilmour, D. (2009). Collective cell migration in morphogenesis,

regeneration and cancer. Nat Rev Mol Cell Biol 10, 445-57.

Friedland, J. C., Lee, M. H. and Boettiger, D. (2009). Mechanically activated integrin

switch controls alpha5beta1 function. Science 323, 642-4.

Funderburgh, J. L. (2002). Keratan sulfate biosynthesis. IUBMB Life 54, 187-94.

Funderburgh, J. L., Mann, M. M. and Funderburgh, M. L. (2003). Keratocyte

phenotype mediates proteoglycan structure: a role for fibroblasts in corneal fibrosis. J Biol Chem
278, 45629-37.

139
 Gargett, C. E., Cornish, E. J. and Wiley, J. S. (1996). Phospholipase D activation by
P2Z-purinoceptor agonists in human lymphocytes is dependent on bivalent cation influx.
Biochem J 313 (Pt 2), 529-35.

Garibaldi, D. C. and Adler, R. A. (2007). Cicatricial ectropion associated with

treatment of metastatic colorectal cancer with cetuximab. Ophthal Plast Reconstr Surg 23, 62-3.

Genetos, D. C., Kephart, C. J., Zhang, Y., Yellowley, C. E. and Donahue, H. J.

(2007). Oscillating fluid flow activation of gap junction hemichannels induces atp release from
MLO-Y4 osteocytes. J Cell Physiol 212, 207-14.

Geyti, C. S., Odgaard, E., Overgaard, M. T., Jensen, M. E., Leipziger, J. and
Praetorius, H. A. (2008). Slow spontaneous [Ca2+] i oscillations reflect nucleotide release from
renal epithelia. Pflugers Arch 455, 1105-17.

Giannone, G. and Sheetz, M. P. (2006). Substrate rigidity and force define form
through tyrosine phosphatase and kinase pathways. Trends Cell Biol 16, 213-23.

Gipson, I. K. (2004). Distribution of mucins at the ocular surface. Exp Eye Res 78, 379-
88.

Gipson, I. K. (2007). The ocular surface: the challenge to enable and protect vision: the
Friedenwald lecture. Invest Ophthalmol Vis Sci 48, 4390; 4391-8.

Gipson, I. K., Spurr-Michaud, S., Argueso, P., Tisdale, A., Ng, T. F. and Russo, C.
L. (2003). Mucin gene expression in immortalized human corneal-limbal and conjunctival
epithelial cell lines. Invest Ophthalmol Vis Sci 44, 2496-506.

Glading, A., Chang, P., Lauffenburger, D. A. and Wells, A. (2000). Epidermal growth
factor receptor activation of calpain is required for fibroblast motility and occurs via an
ERK/MAP kinase signaling pathway. J Biol Chem 275, 2390-8.

Gomes, P., Srinivas, S. P., Van Driessche, W., Vereecke, J. and Himpens, B. (2005).
ATP release through connexin hemichannels in corneal endothelial cells. Invest Ophthalmol Vis
Sci 46, 1208-18.

Goodenough, D. A. and Paul, D. L. (2003). Beyond the gap: functions of unpaired

connexon channels. Nat Rev Mol Cell Biol 4, 285-94.

Goodlad, R. A. and Wright, N. A. (1995). Epidermal growth factor and transforming

growth factor-alpha actions on the gut. Eur J Gastroenterol Hepatol 7, 928-32.

Gregg, D., Rauscher, F. M. and Goldschmidt-Clermont, P. J. (2003). Rac regulates

cardiovascular superoxide through diverse molecular interactions: more than a binary GTP
switch. Am J Physiol Cell Physiol 285, C723-34.

Grose, R. and Werner, S. (2004). Wound-healing studies in transgenic and knockout

mice. Mol Biotechnol 28, 147-66.

140
 Guo, W. and Giancotti, F. G. (2004). Integrin signalling during tumour progression. Nat
Rev Mol Cell Biol 5, 816-26.

Gurtner, G. C., Werner, S., Barrandon, Y. and Longaker, M. T. (2008). Wound

repair and regeneration. Nature 453, 314-21.

Hagmann, W., Nies, A. T., Konig, J., Frey, M., Zentgraf, H. and Keppler, D. (1999).
Purification of the human apical conjugate export pump MRP2 reconstitution and functional
characterization as substrate-stimulated ATPase. Eur J Biochem 265, 281-9.

Han, B., Bai, X. H., Lodyga, M., Xu, J., Yang, B. B., Keshavjee, S., Post, M. and Liu,
M. (2004). Conversion of mechanical force into biochemical signaling. J Biol Chem 279, 54793-
801.

Hardwicke, J., Ferguson, E. L., Moseley, R., Stephens, P., Thomas, D. W. and
Duncan, R. (2008). Dextrin-rhEGF conjugates as bioresponsive nanomedicines for wound
repair. J Control Release 130, 275-83.

Hardwicke, J., Schmaljohann, D., Boyce, D. and Thomas, D. (2008). Epidermal

growth factor therapy and wound healing--past, present and future perspectives. Surgeon 6, 172-
7.

Harks, E. G., de Roos, A. D., Peters, P. H., de Haan, L. H., Brouwer, A., Ypey, D. L.,
van Zoelen, E. J. and Theuvenet, A. P. (2001). Fenamates: a novel class of reversible gap
junction blockers. J Pharmacol Exp Ther 298, 1033-41.

Harris, R. C., Chung, E. and Coffey, R. J. (2003). EGF receptor ligands. Exp Cell Res
284, 2-13.

He, J. and Bazan, H. E. (2008). Epidermal growth factor synergism with TGF-beta1 via
PI-3 kinase activity in corneal keratocyte differentiation. Invest Ophthalmol Vis Sci 49, 2936-45.

Higashiyama, S., Iwabuki, H., Morimoto, C., Hieda, M., Inoue, H. and Matsushita,
N. (2008). Membrane-anchored growth factors, the epidermal growth factor family: beyond
receptor ligands. Cancer Sci 99, 214-20.

Higashiyama, S. and Nanba, D. (2005). ADAM-mediated ectodomain shedding of HB-

EGF in receptor cross-talk. Biochim Biophys Acta 1751, 110-7.

Higuchi, S., Ohtsu, H., Suzuki, H., Shirai, H., Frank, G. D. and Eguchi, S. (2007).
Angiotensin II signal transduction through the AT1 receptor: novel insights into mechanisms and
pathophysiology. Clin Sci (Lond) 112, 417-28.

Hirotani, S., Higuchi, Y., Nishida, K., Nakayama, H., Yamaguchi, O., Hikoso, S.,
Takeda, T., Kashiwase, K., Watanabe, T., Asahi, M. et al. (2004). Ca(2+)-sensitive tyrosine
kinase Pyk2/CAK beta-dependent signaling is essential for G-protein-coupled receptor agonist-
induced hypertrophy. J Mol Cell Cardiol 36, 799-807.

141
 Hiroyama, M. and Exton, J. H. (2005). Localization and regulation of phospholipase
D2 by ARF6. J Cell Biochem 95, 149-64.

Hori, K., Sotozono, C., Hamuro, J., Yamasaki, K., Kimura, Y., Ozeki, M., Tabata,
Y. and Kinoshita, S. (2007). Controlled-release of epidermal growth factor from cationized
gelatin hydrogel enhances corneal epithelial wound healing. J Control Release 118, 169-76.

Horiuchi, K., Le Gall, S., Schulte, M., Yamaguchi, T., Reiss, K., Murphy, G.,
Toyama, Y., Hartmann, D., Saftig, P. and Blobel, C. P. (2007). Substrate selectivity of
epidermal growth factor-receptor ligand sheddases and their regulation by phorbol esters and
calcium influx. Mol Biol Cell 18, 176-88.

Huang, P. and Frohman, M. A. (2007). The potential for phospholipase D as a new

therapeutic target. Expert Opin Ther Targets 11, 707-16.

Hudson, L. G. and McCawley, L. J. (1998). Contributions of the epidermal growth

factor receptor to keratinocyte motility. Microsc Res Tech 43, 444-55.

Humphries, M. J., McEwan, P. A., Barton, S. J., Buckley, P. A., Bella, J. and Mould,
A. P. (2003). Integrin structure: heady advances in ligand binding, but activation still makes the
knees wobble. Trends Biochem Sci 28, 313-20.

Hynes, R. O. (2002). Integrins: bidirectional, allosteric signaling machines. Cell 110,

673-87.

Idone, V., Tam, C. and Andrews, N. W. (2008). Two-way traffic on the road to plasma
membrane repair. Trends Cell Biol 18, 552-9.

Iiizumi, M., Bandyopadhyay, S., Pai, S. K., Watabe, M., Hirota, S., Hosobe, S.,
Tsukada, T., Miura, K., Saito, K., Furuta, E. et al. (2008). RhoC promotes metastasis via
activation of the Pyk2 pathway in prostate cancer. Cancer Res 68, 7613-20.

Imanishi, J., Kamiyama, K., Iguchi, I., Kita, M., Sotozono, C. and Kinoshita, S.
(2000). Growth factors: importance in wound healing and maintenance of transparency of the
cornea. Prog Retin Eye Res 19, 113-29.

Ingber, D. E. (2006). Cellular mechanotransduction: putting all the pieces together again.
Faseb J 20, 811-27.

Ingley, E. (2008). Src family kinases: regulation of their activities, levels and
identification of new pathways. Biochim Biophys Acta 1784, 56-65.

Isakson, B. E., Evans, W. H. and Boitano, S. (2001). Intercellular Ca2+ signaling in

alveolar epithelial cells through gap junctions and by extracellular ATP. Am J Physiol Lung Cell
Mol Physiol 280, L221-8.

Iwamoto, R., Higashiyama, S., Mitamura, T., Taniguchi, N., Klagsbrun, M. and
Mekada, E. (1994). Heparin-binding EGF-like growth factor, which acts as the diphtheria toxin

142
receptor, forms a complex with membrane protein DRAP27/CD9, which up-regulates functional
receptors and diphtheria toxin sensitivity. Embo J 13, 2322-30.

Iwamoto, R. and Mekada, E. (2006). ErbB and HB-EGF signaling in heart development
and function. Cell Struct Funct 31, 1-14.

Iwamoto, R., Yamazaki, S., Asakura, M., Takashima, S., Hasuwa, H., Miyado, K.,
Adachi, S., Kitakaze, M., Hashimoto, K., Raab, G. et al. (2003). Heparin-binding EGF-like
growth factor and ErbB signaling is essential for heart function. Proc Natl Acad Sci U S A 100,
3221-6.

Iwasaki, H., Eguchi, S., Ueno, H., Marumo, F. and Hirata, Y. (2000). Mechanical
stretch stimulates growth of vascular smooth muscle cells via epidermal growth factor receptor.
Am J Physiol Heart Circ Physiol 278, H521-9.

Iyer, A. K., Tran, K. T., Griffith, L. and Wells, A. (2008). Cell surface restriction of
EGFR by a tenascin cytotactin-encoded EGF-like repeat is preferential for motility-related
signaling. J Cell Physiol 214, 504-12.

Jacinto, A., Martinez-Arias, A. and Martin, P. (2001). Mechanisms of epithelial fusion

and repair. Nat Cell Biol 3, E117-23.

Jackson, L. F., Qiu, T. H., Sunnarborg, S. W., Chang, A., Zhang, C., Patterson, C.
and Lee, D. C. (2003). Defective valvulogenesis in HB-EGF and TACE-null mice is associated
with aberrant BMP signaling. Embo J 22, 2704-16.

Jaffe, L. F. (2007). Stretch-activated calcium channels relay fast calcium waves

propagated by calcium-induced calcium influx. Biol Cell 99, 175-84.

Jane, S. M., Ting, S. B. and Cunningham, J. M. (2005). Epidermal impermeable

barriers in mouse and fly. Curr Opin Genet Dev 15, 447-53.

Jenkins, G. M. and Frohman, M. A. (2005). Phospholipase D: a lipid centric review.

Cell Mol Life Sci 62, 2305-16.

Johnson, K. S., Levin, F. and Chu, D. S. (2009). Persistent corneal epithelial defect
associated with erlotinib treatment. Cornea 28, 706-7.

Jorissen, R. N., Walker, F., Pouliot, N., Garrett, T. P., Ward, C. W. and Burgess, A.
W. (2003). Epidermal growth factor receptor: mechanisms of activation and signalling. Exp Cell
Res 284, 31-53.

Jumblatt, J. E. and Jumblatt, M. M. (1998). Regulation of ocular mucin secretion by

P2Y2 nucleotide receptors in rabbit and human conjunctiva. Exp Eye Res 67, 341-6.

Kagiyama, S., Eguchi, S., Frank, G. D., Inagami, T., Zhang, Y. C. and Phillips, M. I.
(2002). Angiotensin II-induced cardiac hypertrophy and hypertension are attenuated by
epidermal growth factor receptor antisense. Circulation 106, 909-12.

143
 Kahn, C. R., Young, E., Lee, I. H. and Rhim, J. S. (1993). Human corneal epithelial
primary cultures and cell lines with extended life span: in vitro model for ocular studies. Invest
Ophthalmol Vis Sci 34, 3429-41.

Kanno, H., Horikawa, Y., Hodges, R. R., Zoukhri, D., Shatos, M. A., Rios, J. D. and
Dartt, D. A. (2003). Cholinergic agonists transactivate EGFR and stimulate MAPK to induce
goblet cell secretion. Am J Physiol Cell Physiol 284, C988-98.

Kato, T., Saika, S. and Ohnishi, Y. (2006). Effects of the matrix metalloproteinase
inhibitor GM6001 on the destruction and alteration of epithelial basement membrane during the
healing of post-alkali burn in rabbit cornea. Jpn J Ophthalmol 50, 90-5.

Katz, M., Amit, I. and Yarden, Y. (2007). Regulation of MAPKs by growth factors and
receptor tyrosine kinases. Biochim Biophys Acta 1773, 1161-76.

Katzur, A. C., Koshimizu, T., Tomic, M., Schultze-Mosgau, A., Ortmann, O. and
Stojilkovic, S. S. (1999). Expression and responsiveness of P2Y2 receptors in human
endometrial cancer cell lines. J Clin Endocrinol Metab 84, 4085-91.

Kaverina, I., Krylyshkina, O. and Small, J. V. (2002). Regulation of substrate

adhesion dynamics during cell motility. Int J Biochem Cell Biol 34, 746-61.

Kaye, S. and Choudhary, A. (2006). Herpes simplex keratitis. Prog Retin Eye Res 25,
355-80.

Kim, J. H., Kim, H. W., Jeon, H., Suh, P. G. and Ryu, S. H. (2006). Phospholipase D1
regulates cell migration in a lipase activity-independent manner. J Biol Chem 281, 15747-56.

Kim, L. C., Song, L. and Haura, E. B. (2009). Src kinases as therapeutic targets for
cancer. Nat Rev Clin Oncol 6, 587-95.

Klenkler, B. and Sheardown, H. (2004). Growth factors in the anterior segment: role in
tissue maintenance, wound healing and ocular pathology. Exp Eye Res 79, 677-88.

Klenkler, B., Sheardown, H. and Jones, L. (2007). Growth factors in the tear film: role
in tissue maintenance, wound healing, and ocular pathology. Ocul Surf 5, 228-39.

Klepeis, V. E., Cornell-Bell, A. and Trinkaus-Randall, V. (2001). Growth factors but

not gap junctions play a role in injury-induced Ca2+ waves in epithelial cells. J Cell Sci 114,
4185-95.

Klepeis, V. E., Weinger, I., Kaczmarek, E. and Trinkaus-Randall, V. (2004). P2Y

receptors play a critical role in epithelial cell communication and migration. J Cell Biochem 93,
1115-33.

Knight, G. E., Bodin, P., De Groat, W. C. and Burnstock, G. (2002). ATP is released
from guinea pig ureter epithelium on distension. Am J Physiol Renal Physiol 282, F281-8.

144
 Kodama, H., Fukuda, K., Takahashi, T., Sano, M., Kato, T., Tahara, S., Hakuno, D.,
Sato, T., Manabe, T., Konishi, F. et al. (2002). Role of EGF Receptor and Pyk2 in endothelin-
1-induced ERK activation in rat cardiomyocytes. J Mol Cell Cardiol 34, 139-50.

Kohno, T., Matsuda, E., Sasaki, H. and Sasaki, T. (2008). Protein-tyrosine kinase
CAKbeta/PYK2 is activated by binding Ca2+/calmodulin to FERM F2 alpha2 helix and thus
forming its dimer. Biochem J 410, 513-23.

Kuo, I. C. (2004). Corneal wound healing. Curr Opin Ophthalmol 15, 311-5.

Kusner, D. J. and Adams, J. (2000). ATP-induced killing of virulent Mycobacterium

tuberculosis within human macrophages requires phospholipase D. J Immunol 164, 379-88.

Kusner, D. J., Barton, J. A., Wen, K. K., Wang, X., Rubenstein, P. A. and Iyer, S. S.
(2002). Regulation of phospholipase D activity by actin. Actin exerts bidirectional modulation of
Mammalian phospholipase D activity in a polymerization-dependent, isoform-specific manner. J
Biol Chem 277, 50683-92.

Kuwabara, K., Nakaoka, T., Sato, K., Nishishita, T., Sasaki, T. and Yamashita, N.
(2004). Differential regulation of cell migration and proliferation through proline-rich tyrosine
kinase 2 in endothelial cells. Endocrinology 145, 3324-30.

Kuwada, S. K. and Li, X. (2000). Integrin alpha5/beta1 mediates fibronectin-dependent

epithelial cell proliferation through epidermal growth factor receptor activation. Mol Biol Cell
11, 2485-96.

Kwak, B. R. and Jongsma, H. J. (1999). Selective inhibition of gap junction channel

activity by synthetic peptides. J Physiol 516 (Pt 3), 679-85.

Lauffenburger, D. A. and Horwitz, A. F. (1996). Cell migration: a physically

integrated molecular process. Cell 84, 359-69.

Lauffenburger, D. A., Oehrtman, G. T., Walker, L. and Wiley, H. S. (1998). Real-

time quantitative measurement of autocrine ligand binding indicates that autocrine loops are
spatially localized. Proc Natl Acad Sci U S A 95, 15368-73.

Lavieri, R., Scott, S. A., Lewis, J. A., Selvy, P. E., Armstrong, M. D., Alex Brown, H.
and Lindsley, C. W. (2009). Design and synthesis of isoform-selective phospholipase D (PLD)
inhibitors. Part II. Identification of the 1,3,8-triazaspiro[4,5]decan-4-one privileged structure that
engenders PLD2 selectivity. Bioorg Med Chem Lett 19, 2240-3.

Lazarowski, E. R., Boucher, R. C. and Harden, T. K. (2003). Mechanisms of release

of nucleotides and integration of their action as P2X- and P2Y-receptor activating molecules.
Mol Pharmacol 64, 785-95.

Lee, C. S., Kim, I. S., Park, J. B., Lee, M. N., Lee, H. Y., Suh, P. G. and Ryu, S. H.
(2006). The phox homology domain of phospholipase D activates dynamin GTPase activity and
accelerates EGFR endocytosis. Nat Cell Biol 8, 477-84.

145
 Lee, J., Ishihara, A., Oxford, G., Johnson, B. and Jacobson, K. (1999). Regulation of
cell movement is mediated by stretch-activated calcium channels. Nature 400, 382-6.

Lehman, N., Ledford, B., Di Fulvio, M., Frondorf, K., McPhail, L. C. and Gomez-
Cambronero, J. (2007). Phospholipase D2-derived phosphatidic acid binds to and activates
ribosomal p70 S6 kinase independently of mTOR. Faseb J 21, 1075-87.

Lewis, J. A., Scott, S. A., Lavieri, R., Buck, J. R., Selvy, P. E., Stoops, S. L.,
Armstrong, M. D., Brown, H. A. and Lindsley, C. W. (2009). Design and synthesis of
isoform-selective phospholipase D (PLD) inhibitors. Part I: Impact of alternative halogenated
privileged structures for PLD1 specificity. Bioorg Med Chem Lett 19, 1916-20.

Leybaert, L., Braet, K., Vandamme, W., Cabooter, L., Martin, P. E. and Evans, W.
H. (2003). Connexin channels, connexin mimetic peptides and ATP release. Cell Commun Adhes
10, 251-7.

Lezama, R., Diaz-Tellez, A., Ramos-Mandujano, G., Oropeza, L. and Pasantes-

Morales, H. (2005). Epidermal growth factor receptor is a common element in the signaling
pathways activated by cell volume changes in isosmotic, hyposmotic or hyperosmotic
conditions. Neurochem Res 30, 1589-97.

Li, F. and Malik, K. U. (2005). Angiotensin II-induced Akt activation through the
epidermal growth factor receptor in vascular smooth muscle cells is mediated by phospholipid
metabolites derived by activation of phospholipase D. J Pharmacol Exp Ther 312, 1043-54.

Li, H., Liu, T. F., Lazrak, A., Peracchia, C., Goldberg, G. S., Lampe, P. D. and
Johnson, R. G. (1996). Properties and regulation of gap junctional hemichannels in the plasma
membranes of cultured cells. J Cell Biol 134, 1019-30.

Li, X. and Earp, H. S. (1997). Paxillin is tyrosine-phosphorylated by and preferentially

associates with the calcium-dependent tyrosine kinase in rat liver epithelial cells. J Biol Chem
272, 14341-8.

Li, Y., Ren, J., Yu, W., Li, Q., Kuwahara, H., Yin, L., Carraway, K. L., 3rd and
Kufe, D. (2001). The epidermal growth factor receptor regulates interaction of the human
DF3/MUC1 carcinoma antigen with c-Src and beta-catenin. J Biol Chem 276, 35239-42.

Liang, C. C., Park, A. Y. and Guan, J. L. (2007). In vitro scratch assay: a convenient
and inexpensive method for analysis of cell migration in vitro. Nat Protoc 2, 329-33.

Lipinski, C. A., Tran, N. L., Dooley, A., Pang, Y. P., Rohl, C., Kloss, J., Yang, Z.,
McDonough, W., Craig, D., Berens, M. E. et al. (2006). Critical role of the FERM domain in
Pyk2 stimulated glioma cell migration. Biochem Biophys Res Commun 349, 939-47.

Lipinski, C. A., Tran, N. L., Menashi, E., Rohl, C., Kloss, J., Bay, R. C., Berens, M.
E. and Loftus, J. C. (2005). The tyrosine kinase pyk2 promotes migration and invasion of
glioma cells. Neoplasia 7, 435-45.

146
 Liu, Y., Geisbuhler, B. and Jones, A. W. (1992). Activation of multiple mechanisms
including phospholipase D by endothelin-1 in rat aorta. Am J Physiol 262, C941-9.

Loo, D. T. and Rillema, J. R. (1998). Measurement of cell death. Methods Cell Biol 57,
251-64.

Lu, L., Reinach, P. S. and Kao, W. W. (2001). Corneal epithelial wound healing. Exp
Biol Med (Maywood) 226, 653-64.

Luo, B. H., Carman, C. V. and Springer, T. A. (2007). Structural basis of integrin

regulation and signaling. Annu Rev Immunol 25, 619-47.

Lyons, P. D., Dunty, J. M., Schaefer, E. M. and Schaller, M. D. (2001). Inhibition of

the catalytic activity of cell adhesion kinase beta by protein-tyrosine phosphatase-PEST-
mediated dephosphorylation. J Biol Chem 276, 24422-31.

Maroto, R. and Hamill, O. P. (2001). Brefeldin A block of integrin-dependent

mechanosensitive ATP release from Xenopus oocytes reveals a novel mechanism of
mechanotransduction. J Biol Chem 276, 23867-72.

Maroto, R., Raso, A., Wood, T. G., Kurosky, A., Martinac, B. and Hamill, O. P.
(2005). TRPC1 forms the stretch-activated cation channel in vertebrate cells. Nat Cell Biol 7,
179-85.

Martin, P. (1997). Wound healing--aiming for perfect skin regeneration. Science 276,
75-81.

Martin, P. and Parkhurst, S. M. (2004). Parallels between tissue repair and embryo
morphogenesis. Development 131, 3021-34.

Martin, T. W., Feldman, D. R., Goldstein, K. E. and Wagner, J. R. (1989). Long-term

phorbol ester treatment dissociates phospholipase D activation from phosphoinositide hydrolysis
and prostacyclin synthesis in endothelial cells stimulated with bradykinin. Biochem Biophys Res
Commun 165, 319-26.

Mazie, A. R., Spix, J. K., Block, E. R., Achebe, H. B. and Klarlund, J. K. (2006).
Epithelial cell motility is triggered by activation of the EGF receptor through phosphatidic acid
signaling. J Cell Sci 119, 1645-54.

McDermott, M., Wakelam, M. J. and Morris, A. J. (2004). Phospholipase D. Biochem

Cell Biol 82, 225-53.

McKay, M. M. and Morrison, D. K. (2007). Integrating signals from RTKs to

ERK/MAPK. Oncogene 26, 3113-21.

McNeil, P. L. and Kirchhausen, T. (2005). An emergency response team for membrane

repair. Nat Rev Mol Cell Biol 6, 499-505.

147
 Menashi, E. B. and Loftus, J. C. (2009). Differential effects of Pyk2 and FAK on the
hypertrophic response of cardiac myocytes. Cell Tissue Res 337, 243-55.

Mine, N., Iwamoto, R. and Mekada, E. (2005). HB-EGF promotes epithelial cell
migration in eyelid development. Development 132, 4317-26.

Mitchison, T. J. and Cramer, L. P. (1996). Actin-based cell motility and cell

locomotion. Cell 84, 371-9.

Mitra, S. K. and Schlaepfer, D. D. (2006). Integrin-regulated FAK-Src signaling in

normal and cancer cells. Curr Opin Cell Biol 18, 516-23.

Mohan, R. R., Sharma, A., Netto, M. V., Sinha, S. and Wilson, S. E. (2005). Gene
therapy in the cornea. Prog Retin Eye Res 24, 537-59.

Montiel, M., Quesada, J. and Jimenez, E. (2007). Activation of calcium-dependent

kinases and epidermal growth factor receptor regulate muscarinic acetylcholine receptor-
mediated MAPK/ERK activation in thyroid epithelial cells. Cell Signal 19, 2138-46.

Moro, L., Dolce, L., Cabodi, S., Bergatto, E., Boeri Erba, E., Smeriglio, M., Turco,
E., Retta, S. F., Giuffrida, M. G., Venturino, M. et al. (2002). Integrin-induced epidermal
growth factor (EGF) receptor activation requires c-Src and p130Cas and leads to phosphorylation
of specific EGF receptor tyrosines. J Biol Chem 277, 9405-14.

Moro, L., Venturino, M., Bozzo, C., Silengo, L., Altruda, F., Beguinot, L., Tarone,
G. and Defilippi, P. (1998). Integrins induce activation of EGF receptor: role in MAP kinase
induction and adhesion-dependent cell survival. Embo J 17, 6622-32.

Morris, A. J. (2007). Regulation of phospholipase D activity, membrane targeting and

intracellular trafficking by phosphoinositides. Biochem Soc Symp, 247-57.

Myhre, G. M., Toruner, M., Abraham, S. and Egan, L. J. (2004). Metalloprotease

disintegrin-mediated ectodomain shedding of EGFR ligands promotes intestinal epithelial
restitution. Am J Physiol Gastrointest Liver Physiol 287, G1213-9.

Na, S., Collin, O., Chowdhury, F., Tay, B., Ouyang, M., Wang, Y. and Wang, N.
(2008). Rapid signal transduction in living cells is a unique feature of mechanotransduction.
Proc Natl Acad Sci U S A 105, 6626-31.

Nakamura, Y., Sotozono, C. and Kinoshita, S. (2001). The epidermal growth factor
receptor (EGFR): role in corneal wound healing and homeostasis. Exp Eye Res 72, 511-7.

Nakashima, I., Takeda, K., Kawamoto, Y., Okuno, Y., Kato, M. and Suzuki, H.
(2005). Redox control of catalytic activities of membrane-associated protein tyrosine kinases.
Arch Biochem Biophys 434, 3-10.

148
 Netto, M. V., Mohan, R. R., Ambrosio, R., Jr., Hutcheon, A. E., Zieske, J. D. and
Wilson, S. E. (2005). Wound healing in the cornea: a review of refractive surgery complications
and new prospects for therapy. Cornea 24, 509-22.

Netto, M. V., Mohan, R. R., Sinha, S., Sharma, A., Dupps, W. and Wilson, S. E.
(2006). Stromal haze, myofibroblasts, and surface irregularity after PRK. Exp Eye Res 82, 788-
97.

Nikolic, D. L., Boettiger, A. N., Bar-Sagi, D., Carbeck, J. D. and Shvartsman, S. Y.

(2006). Role of boundary conditions in an experimental model of epithelial wound healing. Am J
Physiol Cell Physiol 291, C68-75.

Nishida, T. and Yanai, R. (2009). Advances in treatment for neurotrophic keratopathy.

Curr Opin Ophthalmol 20, 276-81.

Nobes, C. D. and Hall, A. (1999). Rho GTPases control polarity, protrusion, and
adhesion during cell movement. J Cell Biol 144, 1235-44.

Obata, H. and Tsuru, T. (2007). Corneal wound healing from the perspective of
keratoplasty specimens with special reference to the function of the Bowman layer and Descemet
membrane. Cornea 26, S82-9.

Oehrtman, G. T., Wiley, H. S. and Lauffenburger, D. A. (1998). Escape of autocrine

ligands into extracellular medium: experimental test of theoretical model predictions. Biotechnol
Bioeng 57, 571-82.

Ohtsu, H., Dempsey, P. J. and Eguchi, S. (2006). ADAMs as mediators of EGF

receptor transactivation by G protein-coupled receptors. Am J Physiol Cell Physiol 291, C1-10.

Okigaki, M., Davis, C., Falasca, M., Harroch, S., Felsenfeld, D. P., Sheetz, M. P. and
Schlessinger, J. (2003). Pyk2 regulates multiple signaling events crucial for macrophage
morphology and migration. Proc Natl Acad Sci U S A 100, 10740-5.

Omelchenko, T., Vasiliev, J. M., Gelfand, I. M., Feder, H. H. and Bonder, E. M.

(2003). Rho-dependent formation of epithelial "leader" cells during wound healing. Proc Natl
Acad Sci U S A 100, 10788-93.

Orr, A. W., Helmke, B. P., Blackman, B. R. and Schwartz, M. A. (2006). Mechanisms

of mechanotransduction. Dev Cell 10, 11-20.

Oude Weernink, P. A., Lopez de Jesus, M. and Schmidt, M. (2007). Phospholipase D

signaling: orchestration by PIP2 and small GTPases. Naunyn Schmiedebergs Arch Pharmacol
374, 399-411.

Owen, K. A., Pixley, F. J., Thomas, K. S., Vicente-Manzanares, M., Ray, B. J.,
Horwitz, A. F., Parsons, J. T., Beggs, H. E., Stanley, E. R. and Bouton, A. H. (2007).
Regulation of lamellipodial persistence, adhesion turnover, and motility in macrophages by focal
adhesion kinase. J Cell Biol 179, 1275-87.

149
 Panchin, Y., Kelmanson, I., Matz, M., Lukyanov, K., Usman, N. and Lukyanov, S.
(2000). A ubiquitous family of putative gap junction molecules. Curr Biol 10, R473-4.

Park, S. Y., Li, H. and Avraham, S. (2007). RAFTK/Pyk2 regulates EGF-induced

PC12 cell spreading and movement. Cell Signal 19, 289-300.

Park, S. Y., Schinkmann, K. A. and Avraham, S. (2006). RAFTK/Pyk2 mediates

LPA-induced PC12 cell migration. Cell Signal 18, 1063-71.

Parmentier, J. H., Pavicevic, Z. and Malik, K. U. (2006). ANG II stimulates

phospholipase D through PKCzeta activation in VSMC: implications in adhesion, spreading, and
hypertrophy. Am J Physiol Heart Circ Physiol 290, H46-54.

Partridge, E. A., Le Roy, C., Di Guglielmo, G. M., Pawling, J., Cheung, P.,
Granovsky, M., Nabi, I. R., Wrana, J. L. and Dennis, J. W. (2004). Regulation of cytokine
receptors by Golgi N-glycan processing and endocytosis. Science 306, 120-4.

Pastor, J. C. and Calonge, M. (1992). Epidermal growth factor and corneal wound
healing. A multicenter study. Cornea 11, 311-4.

Pei, Y. and Tuschl, T. (2006). On the art of identifying effective and specific siRNAs.
Nat Methods 3, 670-6.

Perez-Andres, E., Fernandez-Rodriguez, M., Gonzalez, M., Zubiaga, A., Vallejo, A.,
Garcia, I., Matute, C., Pochet, S., Dehaye, J. P., Trueba, M. et al. (2002). Activation of
phospholipase D-2 by P2X(7) agonists in rat submandibular gland acini. J Lipid Res 43, 1244-
55.

Peus, D., Vasa, R. A., Meves, A., Pott, M., Beyerle, A., Squillace, K. and Pittelkow,
M. R. (1998). H2O2 is an important mediator of UVB-induced EGF-receptor phosphorylation in
cultured keratinocytes. J Invest Dermatol 110, 966-71.

Playford, M. P. and Schaller, M. D. (2004). The interplay between Src and integrins in
normal and tumor biology. Oncogene 23, 7928-46.

Pochet, S., Gomez-Munoz, A., Marino, A. and Dehaye, J. P. (2003). Regulation of

phospholipase D by P2X7 receptors in submandibular ductal cells. Cell Signal 15, 927-35.

Poujade, M., Grasland-Mongrain, E., Hertzog, A., Jouanneau, J., Chavrier, P.,
Ladoux, B., Buguin, A. and Silberzan, P. (2007). Collective migration of an epithelial
monolayer in response to a model wound. Proc Natl Acad Sci U S A 104, 15988-93.

Praetorius, H. A. and Leipziger, J. (2009). ATP release from non-excitable cells.

Purinergic Signal 5, 433-46.

Preininger, A. M., Henage, L. G., Oldham, W. M., Yoon, E. J., Hamm, H. E. and
Brown, H. A. (2006). Direct modulation of phospholipase D activity by Gbetagamma. Mol
Pharmacol 70, 311-8.

150
 Prenzel, N., Zwick, E., Leserer, M. and Ullrich, A. (2000). Tyrosine kinase signalling
in breast cancer. Epidermal growth factor receptor: convergence point for signal integration and
diversification. Breast Cancer Res 2, 184-90.

Puddicombe, S. M., Polosa, R., Richter, A., Krishna, M. T., Howarth, P. H., Holgate,
S. T. and Davies, D. E. (2000). Involvement of the epidermal growth factor receptor in
epithelial repair in asthma. Faseb J 14, 1362-74.

Purkiss, J., Owen, P. J., Jones, J. A. and Boarder, M. R. (1992). Stimulation of

phosphatidic acid synthesis in bovine aortic endothelial cells in response to activation of P2-
purinergic receptors. Biochem Pharmacol 43, 1235-42.

Pushker, N., Dada, T., Vajpayee, R. B., Gupta, V., Aggrawal, T. and Titiyal, J. S.
(2001). Neurotrophic keratopathy. Clao J 27, 100-7.

Rand, H. W. (1915). Wound closure in actinian tentacles with reference to the problem
of organization. Rouxs Arch. Entwicklungsmech. Organismen 41, 159-214.

Rao, J. N., Li, L., Golovina, V. A., Platoshyn, O., Strauch, E. D., Yuan, J. X. and
Wang, J. Y. (2001). Ca2+-RhoA signaling pathway required for polyamine-dependent intestinal
epithelial cell migration. Am J Physiol Cell Physiol 280, C993-1007.

Rao, J. N., Platoshyn, O., Golovina, V. A., Liu, L., Zou, T., Marasa, B. S., Turner, D.
J., Yuan, J. X. and Wang, J. Y. (2006). TRPC1 functions as a store-operated Ca2+ channel in
intestinal epithelial cells and regulates early mucosal restitution after wounding. Am J Physiol
Gastrointest Liver Physiol 290, G782-92.

Reddy, C. C., Niyogi, S. K., Wells, A., Wiley, H. S. and Lauffenburger, D. A. (1996).
Engineering epidermal growth factor for enhanced mitogenic potency. Nat Biotechnol 14, 1696-
9.

Reichelt, J. (2007). Mechanotransduction of keratinocytes in culture and in the

epidermis. Eur J Cell Biol.

Repertinger, S. K., Campagnaro, E., Fuhrman, J., El-Abaseri, T., Yuspa, S. H. and
Hansen, L. A. (2004). EGFR enhances early healing after cutaneous incisional wounding. J
Invest Dermatol 123, 982-9.

Revoltella, R. P., Papini, S., Rosellini, A. and Michelini, M. (2007). Epithelial stem
cells of the eye surface. Cell Prolif 40, 445-61.

Rhee, S. G. (2006). Cell signaling. H2O2, a necessary evil for cell signaling. Science
312, 1882-3.

Rhee, S. G., Kang, S. W., Jeong, W., Chang, T. S., Yang, K. S. and Woo, H. A.
(2005). Intracellular messenger function of hydrogen peroxide and its regulation by
peroxiredoxins. Curr Opin Cell Biol 17, 183-9.

151
 Roelle, S., Grosse, R., Buech, T., Chubanov, V. and Gudermann, T. (2008). Essential
role of Pyk2 and Src kinase activation in neuropeptide-induced proliferation of small cell lung
cancer cells. Oncogene 27, 1737-48.

Roskoski, R., Jr. (2005). Src kinase regulation by phosphorylation and

dephosphorylation. Biochem Biophys Res Commun 331, 1-14.

Rubinfeld, H. and Seger, R. (2005). The ERK cascade: a prototype of MAPK signaling.
Mol Biotechnol 31, 151-74.

Saika, S. (2004). TGF-beta signal transduction in corneal wound healing as a therapeutic

target. Cornea 23, S25-30.

Saika, S., Kobata, S., Hashizume, N., Okada, Y. and Yamanaka, O. (1993). Epithelial
basement membrane in alkali-burned corneas in rats. Immunohistochemical study. Cornea 12,
383-90.

Sammak, P. J., Hinman, L. E., Tran, P. O., Sjaastad, M. D. and Machen, T. E.

(1997). How do injured cells communicate with the surviving cell monolayer? J Cell Sci 110 (Pt
4), 465-75.

Sanderson, M. P., Dempsey, P. J. and Dunbar, A. J. (2006). Control of ErbB signaling

through metalloprotease mediated ectodomain shedding of EGF-like factors. Growth Factors 24,
121-36.

Sariola, H. and Saarma, M. (2003). Novel functions and signalling pathways for
GDNF. J Cell Sci 116, 3855-62.

Schaller, M. D. (2008). Calcium-dependent Pyk2 activation: a role for calmodulin?

Biochem J 410, e3-4.

Schallhorn, S. C., Amesbury, E. C. and Tanzer, D. J. (2006). Avoidance, recognition,

and management of LASIK complications. Am J Ophthalmol 141, 733-9.

Schauwienold, D., Sastre, A. P., Genzel, N., Schaefer, M. and Reusch, H. P. (2008).
The transactivated epidermal growth factor receptor recruits Pyk2 to regulate Src kinase activity.
J Biol Chem 283, 27748-56.

Schlaepfer, D. D., Hauck, C. R. and Sieg, D. J. (1999). Signaling through focal

adhesion kinase. Prog Biophys Mol Biol 71, 435-78.

Schugar, R. C., Chirieleison, S. M., Wescoe, K. E., Schmidt, B. T., Askew, Y., Nance,
J. J., Evron, J. M., Peault, B. and Deasy, B. M. (2009). High harvest yield, high expansion,
and phenotype stability of CD146 mesenchymal stromal cells from whole primitive human
umbilical cord tissue. J Biomed Biotechnol 2009, 789526.

Schwiebert, E. M. and Zsembery, A. (2003). Extracellular ATP as a signaling molecule

for epithelial cells. Biochim Biophys Acta 1615, 7-32.

152
 Scott, S. A., Selvy, P. E., Buck, J. R., Cho, H. P., Criswell, T. L., Thomas, A. L.,
Armstrong, M. D., Arteaga, C. L., Lindsley, C. W. and Brown, H. A. (2009). Design of
isoform-selective phospholipase D inhibitors that modulate cancer cell invasiveness. Nat Chem
Biol 5, 108-17.

Shattil, S. J. and Newman, P. J. (2004). Integrins: dynamic scaffolds for adhesion and
signaling in platelets. Blood 104, 1606-15.

Sheetz, M. P. and Dai, J. (1996). Modulation of membrane dynamics and cell motility
by membrane tension. Trends Cell Biol 6, 85-9.

Shell, J. W. (1982). Pharmacokinetics of topically applied ophthalmic drugs. Surv

Ophthalmol 26, 207-18.

Sherratt, J. A. and Dallon, J. C. (2002). Theoretical models of wound healing: past

successes and future challenges. C R Biol 325, 557-64.

Shi, C. S. and Kehrl, J. H. (2004). Pyk2 amplifies epidermal growth factor and c-Src-
induced Stat3 activation. J Biol Chem 279, 17224-31.

Shirakata, Y., Kimura, R., Nanba, D., Iwamoto, R., Tokumaru, S., Morimoto, C.,
Yokota, K., Nakamura, M., Sayama, K., Mekada, E. et al. (2005). Heparin-binding EGF-like
growth factor accelerates keratinocyte migration and skin wound healing. J Cell Sci 118, 2363-
70.

Shurman, D. L., Glazewski, L., Gumpert, A., Zieske, J. D. and Richard, G. (2005).
In vivo and in vitro expression of connexins in the human corneal epithelium. Invest Ophthalmol
Vis Sci 46, 1957-65.

Shvartsman, S. Y., Hagan, M. P., Yacoub, A., Dent, P., Wiley, H. S. and
Lauffenburger, D. A. (2002). Autocrine loops with positive feedback enable context-dependent
cell signaling. Am J Physiol Cell Physiol 282, C545-59.

Sieg, D. J., Ilic, D., Jones, K. C., Damsky, C. H., Hunter, T. and Schlaepfer, D. D.
(1998). Pyk2 and Src-family protein-tyrosine kinases compensate for the loss of FAK in
fibronectin-stimulated signaling events but Pyk2 does not fully function to enhance FAK- cell
migration. Embo J 17, 5933-47.

Smith, N. J., Chan, H. W., Osborne, J. E., Thomas, W. G. and Hannan, R. D. (2004).
Hijacking epidermal growth factor receptors by angiotensin II: new possibilities for
understanding and treating cardiac hypertrophy. Cell Mol Life Sci 61, 2695-703.

Soe, N. N., Ishida, T., Ishida, M., Sawano, M., Abe, K., Miho, N., Chayama, K.,
Kihara, Y. and Yoshizumi, M. (2009). Nifedipine interferes with migration of vascular smooth
muscle cells via inhibition of Pyk2-Src axis. J Atheroscler Thromb 16, 230-8.

153
 Spix, J. K., Chay, E. Y., Block, E. R. and Klarlund, J. K. (2007). Hepatocyte growth
factor induces epithelial cell motility through transactivation of the epidermal growth factor
receptor. Exp Cell Res.

Sponsel, H. T., Breckon, R. and Anderson, R. J. (1995). Adenine nucleotide and

protein kinase C regulation of renal tubular epithelial cell wound healing. Kidney Int 48, 85-92.

Spray, D. C., Ye, Z. C. and Ransom, B. R. (2006). Functional connexin

"hemichannels": a critical appraisal. Glia 54, 758-73.

Stepp, M. A. (2006). Corneal integrins and their functions. Exp Eye Res 83, 3-15.

Stewart, W. W. (1978). Functional connections between cells as revealed by dye-

coupling with a highly fluorescent naphthalimide tracer. Cell 14, 741-59.

Stoscheck, C. M., Nanney, L. B. and King, L. E., Jr. (1992). Quantitative

determination of EGF-R during epidermal wound healing. J Invest Dermatol 99, 645-9.

Stout, C. E., Costantin, J. L., Naus, C. C. and Charles, A. C. (2002). Intercellular

calcium signaling in astrocytes via ATP release through connexin hemichannels. J Biol Chem
277, 10482-8.

Sun, C. K., Man, K., Ng, K. T., Ho, J. W., Lim, Z. X., Cheng, Q., Lo, C. M., Poon, R.
T. and Fan, S. T. (2008). Proline-rich tyrosine kinase 2 (Pyk2) promotes proliferation and
invasiveness of hepatocellular carcinoma cells through c-Src/ERK activation. Carcinogenesis 29,
2096-105.

Sun, S. H., Lin, L. B., Hung, A. C. and Kuo, J. S. (1999). ATP-stimulated Ca2+ influx
and phospholipase D activities of a rat brain-derived type-2 astrocyte cell line, RBA-2, are
mediated through P2X7 receptors. J Neurochem 73, 334-43.

Tamada, M., Perez, T. D., Nelson, W. J. and Sheetz, M. P. (2007). Two distinct modes
of myosin assembly and dynamics during epithelial wound closure. J Cell Biol 176, 27-33.

Tamada, M., Sheetz, M. P. and Sawada, Y. (2004). Activation of a signaling cascade

by cytoskeleton stretch. Dev Cell 7, 709-18.

Thiery, J. P. and Sleeman, J. P. (2006). Complex networks orchestrate epithelial-

mesenchymal transitions. Nat Rev Mol Cell Biol 7, 131-42.

Thomas, P. A. (2003). Fungal infections of the cornea. Eye (Lond) 17, 852-62.

Thomas, S. M. and Brugge, J. S. (1997). Cellular functions regulated by Src family

kinases. Annu Rev Cell Dev Biol 13, 513-609.

Thomas, W. G., Brandenburger, Y., Autelitano, D. J., Pham, T., Qian, H. and
Hannan, R. D. (2002). Adenoviral-directed expression of the type 1A angiotensin receptor

154
promotes cardiomyocyte hypertrophy via transactivation of the epidermal growth factor receptor.
Circ Res 90, 135-42.

Tokumaru, S., Higashiyama, S., Endo, T., Nakagawa, T., Miyagawa, J. I.,
Yamamori, K., Hanakawa, Y., Ohmoto, H., Yoshino, K., Shirakata, Y. et al. (2000).
Ectodomain shedding of epidermal growth factor receptor ligands is required for keratinocyte
migration in cutaneous wound healing. J Cell Biol 151, 209-20.

Tomassen, S. F., van der Wijk, T., de Jonge, H. R. and Tilly, B. C. (2004). Activation
of phospholipase D by osmotic cell swelling. FEBS Lett 566, 287-90.

Torchilin, V. P. (2006). Recent approaches to intracellular delivery of drugs and DNA

and organelle targeting. Annu Rev Biomed Eng 8, 343-75.

Touyz, R. M. and Berry, C. (2002). Recent advances in angiotensin II signaling. Braz J

Med Biol Res 35, 1001-15.

Trollinger, D. R., Isseroff, R. R. and Nuccitelli, R. (2002). Calcium channel blockers

inhibit galvanotaxis in human keratinocytes. J Cell Physiol 193, 1-9.

Tschumperlin, D. J. (2004). EGFR autocrine signaling in a compliant interstitial space:

mechanotransduction from the outside in. Cell Cycle 3, 996-7.

Tschumperlin, D. J., Dai, G., Maly, I. V., Kikuchi, T., Laiho, L. H., McVittie, A. K.,
Haley, K. J., Lilly, C. M., So, P. T., Lauffenburger, D. A. et al. (2004). Mechanotransduction
through growth-factor shedding into the extracellular space. Nature 429, 83-6.

Tsubota, K. (1998). Tear dynamics and dry eye. Prog Retin Eye Res 17, 565-96.

van der Bend, R. L., de Widt, J., van Corven, E. J., Moolenaar, W. H. and van
Blitterswijk, W. J. (1992). The biologically active phospholipid, lysophosphatidic acid, induces
phosphatidylcholine breakdown in fibroblasts via activation of phospholipase D. Comparison
with the response to endothelin. Biochem J 285 (Pt 1), 235-40.

Vermeer, P. D., Einwalter, L. A., Moninger, T. O., Rokhlina, T., Kern, J. A.,
Zabner, J. and Welsh, M. J. (2003). Segregation of receptor and ligand regulates activation of
epithelial growth factor receptor. Nature 422, 322-6.

Wakuta, M., Morishige, N., Chikama, T., Seki, K., Nagano, T. and Nishida, T.
(2007). Delayed wound closure and phenotypic changes in corneal epithelium of the
spontaneously diabetic Goto-Kakizaki rat. Invest Ophthalmol Vis Sci 48, 590-6.

Wang, Z., Haydon, P. G. and Yeung, E. S. (2000). Direct observation of calcium-

independent intercellular ATP signaling in astrocytes. Anal Chem 72, 2001-7.

Warren, C. M. and Landgraf, R. (2006). Signaling through ERBB receptors: multiple

layers of diversity and control. Cell Signal 18, 923-33.

155
 Watanabe, K., Nakagawa, S. and Nishida, T. (1987). Stimulatory effects of fibronectin
and EGF on migration of corneal epithelial cells. Invest Ophthalmol Vis Sci 28, 205-11.

Weber, I. (2006). Is there a pilot in a pseudopod? Eur J Cell Biol 85, 915-24.

Welch, M. D. and Mullins, R. D. (2002). Cellular control of actin nucleation. Annu Rev
Cell Dev Biol 18, 247-88.

Wells, A. (1999). EGF receptor. Int J Biochem Cell Biol 31, 637-43.

Wells, A., Gupta, K., Chang, P., Swindle, S., Glading, A. and Shiraha, H. (1998).
Epidermal growth factor receptor-mediated motility in fibroblasts. Microsc Res Tech 43, 395-
411.

Wesley, U. V., Bove, P. F., Hristova, M., McCarthy, S. and van der Vliet, A. (2007).
Airway epithelial cell migration and wound repair by ATP-mediated activation of dual oxidase 1.
J Biol Chem 282, 3213-20.

Whitcher, J. P., Srinivasan, M. and Upadhyay, M. P. (2001). Corneal blindness: a

global perspective. Bull World Health Organ 79, 214-21.

Wiley, H. S. (2003). Trafficking of the ErbB receptors and its influence on signaling. Exp
Cell Res 284, 78-88.

Williams, K. and Watsky, M. (2002). Gap junctional communication in the human

corneal endothelium and epithelium. Curr Eye Res 25, 29-36.

Wilson, S. A. and Last, A. (2004). Management of corneal abrasions. Am Fam

Physician 70, 123-8.

Wilson, S. E., Mohan, R. R., Mohan, R. R., Ambrosio, R., Jr., Hong, J. and Lee, J.
(2001). The corneal wound healing response: cytokine-mediated interaction of the epithelium,
stroma, and inflammatory cells. Prog Retin Eye Res 20, 625-37.

Wilson, S. E., Netto, M. and Ambrosio, R., Jr. (2003). Corneal cells: chatty in
development, homeostasis, wound healing, and disease. Am J Ophthalmol 136, 530-6.

Woolley, K. and Martin, P. (2000). Conserved mechanisms of repair: from damaged

single cells to wounds in multicellular tissues. Bioessays 22, 911-9.

Wu, R. F., Xu, Y. C., Ma, Z., Nwariaku, F. E., Sarosi, G. A., Jr. and Terada, L. S.
(2005). Subcellular targeting of oxidants during endothelial cell migration. J Cell Biol 171, 893-
904.

Xie, H., Pallero, M. A., Gupta, K., Chang, P., Ware, M. F., Witke, W., Kwiatkowski,
D. J., Lauffenburger, D. A., Murphy-Ullrich, J. E. and Wells, A. (1998). EGF receptor
regulation of cell motility: EGF induces disassembly of focal adhesions independently of the
motility-associated PLCgamma signaling pathway. J Cell Sci 111 (Pt 5), 615-24.

156
 Xiong, W. C., Macklem, M. and Parsons, J. T. (1998). Expression and characterization
of splice variants of PYK2, a focal adhesion kinase-related protein. J Cell Sci 111 (Pt 14), 1981-
91.

Xu, K. P., Ding, Y., Ling, J., Dong, Z. and Yu, F. S. (2004). Wound-induced HB-EGF
ectodomain shedding and EGFR activation in corneal epithelial cells. Invest Ophthalmol Vis Sci
45, 813-20.

Xu, K. P., Li, Y., Ljubimov, A. V. and Yu, F. S. (2009). High glucose suppresses
epidermal growth factor receptor/phosphatidylinositol 3-kinase/Akt signaling pathway and
attenuates corneal epithelial wound healing. Diabetes 58, 1077-85.

Xu, K. P., Yin, J. and Yu, F. S. (2006). SRC-family tyrosine kinases in wound- and
ligand-induced epidermal growth factor receptor activation in human corneal epithelial cells.
Invest Ophthalmol Vis Sci 47, 2832-9.

Xue, C., Wyckoff, J., Liang, F., Sidani, M., Violini, S., Tsai, K. L., Zhang, Z. Y.,
Sahai, E., Condeelis, J. and Segall, J. E. (2006). Epidermal growth factor receptor
overexpression results in increased tumor cell motility in vivo coordinately with enhanced
intravasation and metastasis. Cancer Res 66, 192-7.

Yamada, T., Aoyama, Y., Owada, M. K., Kawakatsu, H. and Kitajima, Y. (2000).
Scraped-wounding causes activation and association of C-Src tyrosine kinase with microtubules
in cultured keratinocytes. Cell Struct Funct 25, 351-9.

Yamada, T., Ueda, T., Ugawa, S., Ishida, Y., Imayasu, M., Koyama, S. and Shimada,
S. Functional expression of transient receptor potential vanilloid 3 (TRPV3) in corneal epithelial
cells: involvement in thermosensation and wound healing. Exp Eye Res 90, 121-9.

Yamaguchi, H. and Condeelis, J. (2007). Regulation of the actin cytoskeleton in cancer

cell migration and invasion. Biochim Biophys Acta 1773, 642-52.

Yang, L., Cranson, D. and Trinkaus-Randall, V. (2004). Cellular injury induces

activation of MAPK via P2Y receptors. J Cell Biochem 91, 938-50.

Yarden, Y. and Sliwkowski, M. X. (2001). Untangling the ErbB signalling network. Nat
Rev Mol Cell Biol 2, 127-37.

Yin, J., Xu, K., Zhang, J., Kumar, A. and Yu, F. S. (2007). Wound-induced ATP
release and EGF receptor activation in epithelial cells. J Cell Sci 120, 815-25.

You, L., Ebner, S. and Kruse, F. E. (2001). Glial cell-derived neurotrophic factor
(GDNF)-induced migration and signal transduction in corneal epithelial cells. Invest Ophthalmol
Vis Sci 42, 2496-504.

Yu, F. S., Yin, J., Xu, K. and Huang, J. (2009). Growth factors and corneal epithelial
wound healing. Brain Res Bull.

157
 Zaidel-Bar, R., Ballestrem, C., Kam, Z. and Geiger, B. (2003). Early molecular events
in the assembly of matrix adhesions at the leading edge of migrating cells. J Cell Sci 116, 4605-
13.

Zhang, Q., Thomas, S. M., Xi, S., Smithgall, T. E., Siegfried, J. M., Kamens, J.,
Gooding, W. E. and Grandis, J. R. (2004). SRC family kinases mediate epidermal growth
factor receptor ligand cleavage, proliferation, and invasion of head and neck cancer cells. Cancer
Res 64, 6166-73.

Zhang, X., Gureasko, J., Shen, K., Cole, P. A. and Kuriyan, J. (2006). An allosteric
mechanism for activation of the kinase domain of epidermal growth factor receptor. Cell 125,
1137-49.

Zhang, Z., Oliver, P., Lancaster, J. J., Schwarzenberger, P. O., Joshi, M. S., Cork, J.
and Kolls, J. K. (2001). Reactive oxygen species mediate tumor necrosis factor alpha-
converting, enzyme-dependent ectodomain shedding induced by phorbol myristate acetate.
Faseb J 15, 303-5.

Zhao, C., Du, G., Skowronek, K., Frohman, M. A. and Bar-Sagi, D. (2007).
Phospholipase D2-generated phosphatidic acid couples EGFR stimulation to Ras activation by
Sos. Nat Cell Biol 9, 706-12.

Zhao, H. B., Yu, N. and Fleming, C. R. (2005). Gap junctional hemichannel-mediated

ATP release and hearing controls in the inner ear. Proc Natl Acad Sci U S A 102, 18724-9.

Zhao, M. (2009). Electrical fields in wound healing-An overriding signal that directs cell
migration. Semin Cell Dev Biol 20, 674-82.

Zhao, M., Dick, A., Forrester, J. V. and McCaig, C. D. (1999). Electric field-directed
cell motility involves up-regulated expression and asymmetric redistribution of the epidermal
growth factor receptors and is enhanced by fibronectin and laminin. Mol Biol Cell 10, 1259-76.

Zheng, Y., Rodrik, V., Toschi, A., Shi, M., Hui, L., Shen, Y. and Foster, D. A. (2006).
Phospholipase D couples survival and migration signals in stress response of human cancer cells.
J Biol Chem 281, 15862-8.

Zieske, J. D., Guimaraes, S. R. and Hutcheon, A. E. (2001). Kinetics of keratocyte

proliferation in response to epithelial debridement. Exp Eye Res 72, 33-9.

Zieske, J. D., Takahashi, H., Hutcheon, A. E. and Dalbone, A. C. (2000). Activation

of epidermal growth factor receptor during corneal epithelial migration. Invest Ophthalmol Vis
Sci 41, 1346-55.

Zrihan-Licht, S., Fu, Y., Settleman, J., Schinkmann, K., Shaw, L., Keydar, I.,
Avraham, S. and Avraham, H. (2000). RAFTK/Pyk2 tyrosine kinase mediates the association
of p190 RhoGAP with RasGAP and is involved in breast cancer cell invasion. Oncogene 19,
1318-28.

158
 Zwick, E., Wallasch, C., Daub, H. and Ullrich, A. (1999). Distinct calcium-dependent
pathways of epidermal growth factor receptor transactivation and PYK2 tyrosine
phosphorylation in PC12 cells. J Biol Chem 274, 20989-96.

159